code
stringlengths 81
54k
| code_codestyle
int64 0
721
| style_context
stringlengths 91
41.9k
| style_context_codestyle
int64 0
699
| label
int64 0
1
|
|---|---|---|---|---|
'''simple docstring'''
import string
from math import logaa
def lowerCAmelCase_ ( __A : str , __A : str ):
'''simple docstring'''
snake_case: List[str] = document.translate(
str.maketrans('' , '' , string.punctuation ) ).replace('\n' , '' )
snake_case: List[str] = document_without_punctuation.split(' ' ) # word tokenization
return len([word for word in tokenize_document if word.lower() == term.lower()] )
def lowerCAmelCase_ ( __A : str , __A : str ):
'''simple docstring'''
snake_case: List[Any] = corpus.lower().translate(
str.maketrans('' , '' , string.punctuation ) ) # strip all punctuation and replace it with ''
snake_case: Optional[int] = corpus_without_punctuation.split('\n' )
snake_case: Union[str, Any] = term.lower()
return (len([doc for doc in docs if term in doc] ), len(__A ))
def lowerCAmelCase_ ( __A : int , __A : int , __A : Any=False ):
'''simple docstring'''
if smoothing:
if n == 0:
raise ValueError('log10(0) is undefined.' )
return round(1 + logaa(n / (1 + df) ) , 3 )
if df == 0:
raise ZeroDivisionError('df must be > 0' )
elif n == 0:
raise ValueError('log10(0) is undefined.' )
return round(logaa(n / df ) , 3 )
def lowerCAmelCase_ ( __A : int , __A : int ):
'''simple docstring'''
return round(tf * idf , 3 ) | 692 |
'''simple docstring'''
import math
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or number % 2 == 0 or number % 3 == 0:
# Negatives, 0, 1, all even numbers, all multiples of 3 are not primes
return False
# All primes number are in format of 6k +/- 1
for i in range(5 , int(math.sqrt(__A ) + 1 ) , 6 ):
if number % i == 0 or number % (i + 2) == 0:
return False
return True
def lowerCAmelCase_ ( __A : float = 0.1 ):
'''simple docstring'''
snake_case: Optional[int] = 3
snake_case: int = 3
while primes / (2 * j - 1) >= ratio:
for i in range(j * j + j + 1 , (j + 2) * (j + 2) , j + 1 ):
primes += is_prime(__A )
j += 2
return j
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = 256
# Modulus to hash a string
__UpperCAmelCase = 1_000_003
def lowerCAmelCase_ ( __A : str , __A : str ):
'''simple docstring'''
snake_case: List[Any] = len(__A )
snake_case: Dict = len(__A )
if p_len > t_len:
return False
snake_case: Dict = 0
snake_case: int = 0
snake_case: int = 1
# Calculating the hash of pattern and substring of text
for i in range(__A ):
snake_case: List[Any] = (ord(pattern[i] ) + p_hash * alphabet_size) % modulus
snake_case: Union[str, Any] = (ord(text[i] ) + text_hash * alphabet_size) % modulus
if i == p_len - 1:
continue
snake_case: Any = (modulus_power * alphabet_size) % modulus
for i in range(0 , t_len - p_len + 1 ):
if text_hash == p_hash and text[i : i + p_len] == pattern:
return True
if i == t_len - p_len:
continue
# Calculate the https://en.wikipedia.org/wiki/Rolling_hash
snake_case: Any = (
(text_hash - ord(text[i] ) * modulus_power) * alphabet_size
+ ord(text[i + p_len] )
) % modulus
return False
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = 'abc1abc12'
snake_case: Any = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
snake_case: Union[str, Any] = 'alskfjaldsk23adsfabcabc'
assert rabin_karp(__A , __A ) and not rabin_karp(__A , __A )
# Test 2)
snake_case: List[Any] = 'ABABX'
snake_case: Optional[int] = 'ABABZABABYABABX'
assert rabin_karp(__A , __A )
# Test 3)
snake_case: Tuple = 'AAAB'
snake_case: Any = 'ABAAAAAB'
assert rabin_karp(__A , __A )
# Test 4)
snake_case: str = 'abcdabcy'
snake_case: Optional[Any] = 'abcxabcdabxabcdabcdabcy'
assert rabin_karp(__A , __A )
# Test 5)
snake_case: Any = 'Lü'
snake_case: Optional[int] = 'Lüsai'
assert rabin_karp(__A , __A )
snake_case: Optional[Any] = 'Lue'
assert not rabin_karp(__A , __A )
print('Success.' )
if __name__ == "__main__":
test_rabin_karp() | 692 |
'''simple docstring'''
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCAmelCase = "pt"
elif is_tf_available():
__UpperCAmelCase = "tf"
else:
__UpperCAmelCase = "jax"
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ByTaTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: int = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ByTaTokenizer.from_pretrained('google/byt5-small' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=20 , SCREAMING_SNAKE_CASE__=5 ):
'''simple docstring'''
snake_case: Optional[Any] = []
for i in range(len(SCREAMING_SNAKE_CASE__ ) ):
try:
snake_case: Optional[Any] = tokenizer.decode([i] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
except UnicodeDecodeError:
pass
toks.append((i, tok) )
snake_case: List[str] = list(filter(lambda SCREAMING_SNAKE_CASE__ : re.match(r'^[ a-zA-Z]+$' , t[1] ) , SCREAMING_SNAKE_CASE__ ) )
snake_case: str = list(filter(lambda SCREAMING_SNAKE_CASE__ : [t[0]] == tokenizer.encode(t[1] , add_special_tokens=SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ ) )
if max_length is not None and len(SCREAMING_SNAKE_CASE__ ) > max_length:
snake_case: Union[str, Any] = toks[:max_length]
if min_length is not None and len(SCREAMING_SNAKE_CASE__ ) < min_length and len(SCREAMING_SNAKE_CASE__ ) > 0:
while len(SCREAMING_SNAKE_CASE__ ) < min_length:
snake_case: Tuple = toks + toks
# toks_str = [t[1] for t in toks]
snake_case: Dict = [t[0] for t in toks]
# Ensure consistency
snake_case: int = tokenizer.decode(SCREAMING_SNAKE_CASE__ , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
if " " not in output_txt and len(SCREAMING_SNAKE_CASE__ ) > 1:
snake_case: str = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
)
if with_prefix_space:
snake_case: Tuple = ' ' + output_txt
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
return output_txt, output_ids
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: str = tokenizer(['hi</s>', 'I went to the gym</s>', '</s>'] )
snake_case: List[Any] = tokenizer(['hi', 'I went to the gym', ''] )
self.assertListEqual(batch_with_eos_added['input_ids'] , batch_without_eos_added['input_ids'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: Union[str, Any] = 'Unicode €.'
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [88, 1_13, 1_08, 1_02, 1_14, 1_03, 1_04, 35, 2_29, 1_33, 1_75, 49, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[str] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'Unicode €.</s>' )
snake_case: List[Any] = tokenizer('e è é ê ë' )
snake_case: Optional[Any] = [1_04, 35, 1_98, 1_71, 35, 1_98, 1_72, 35, 1_98, 1_73, 35, 1_98, 1_74, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[Any] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'e è é ê ë</s>' )
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë' ) ) , 'e è é ê ë</s>' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.ta_base_tokenizer
snake_case: Optional[Any] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case: Optional[int] = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 1, 0]
# fmt: on
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if FRAMEWORK != "jax":
snake_case: Optional[Any] = list(batch.input_ids.numpy()[0] )
else:
snake_case: Dict = list(batch.input_ids.tolist()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 37) , batch.input_ids.shape )
self.assertEqual((2, 37) , batch.attention_mask.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.ta_base_tokenizer
snake_case: List[str] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case: Union[str, Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_attention_mask' , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.ta_base_tokenizer
snake_case: str = [
'Summary of the text.',
'Another summary.',
]
snake_case: Dict = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.ta_base_tokenizer
snake_case: Optional[int] = ['A long paragraph for summarization. </s>']
snake_case: str = ['Summary of the text. </s>']
# fmt: off
snake_case: str = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 35, 1]
snake_case: Optional[int] = [86, 1_20, 1_12, 1_12, 1_00, 1_17, 1_24, 35, 1_14, 1_05, 35, 1_19, 1_07, 1_04, 35, 1_19, 1_04, 1_23, 1_19, 49, 35, 1]
# fmt: on
snake_case: List[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , text_target=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['input_ids'][0] )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['labels'][0] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: Optional[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: Union[str, Any] = tempfile.mkdtemp()
snake_case: Dict = ' He is very happy, UNwant\u00E9d,running'
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'] )
snake_case: List[str] = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token' )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: int = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn('new_additional_special_token' , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Union[str, Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: str = json.load(SCREAMING_SNAKE_CASE__ )
snake_case: int = [F"""<extra_id_{i}>""" for i in range(1_25 )]
snake_case: Optional[int] = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case: str = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Dict = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , )
self.assertIn(
'an_additional_special_token' , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['an_additional_special_token'] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'] ) ) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: Union[str, Any] = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token' , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: Union[str, Any] = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , )
self.assertIn('a_new_additional_special_token' , tokenizer.additional_special_tokens )
self.assertEqual(
['a_new_additional_special_token'] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'] ) ) , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertTrue(tokenizer.decode([2_55] ) == '' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_tokenizers(fast=SCREAMING_SNAKE_CASE__ , do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = ['t', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 'x', 't', '</s>']
snake_case: List[str] = tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Optional[Any] = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Dict = 0
snake_case: List[Any] = tokenizer.convert_ids_to_tokens(
SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [token_id_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [token_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [token_id_to_test_setters] ) | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
from collections.abc import Iterable, Iterator
from dataclasses import dataclass
__UpperCAmelCase = (3, 9, -11, 0, 7, 5, 1, -1)
__UpperCAmelCase = (4, 6, 2, 0, 8, 10, 3, -2)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = 42
__UpperCamelCase = 42
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Node | None = None
for i in sorted(SCREAMING_SNAKE_CASE__ , reverse=SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = Node(SCREAMING_SNAKE_CASE__ , self.head )
def __iter__( self ):
'''simple docstring'''
snake_case: str = self.head
while node:
yield node.data
snake_case: Dict = node.next_node
def __len__( self ):
'''simple docstring'''
return sum(1 for _ in self )
def __str__( self ):
'''simple docstring'''
return " -> ".join([str(SCREAMING_SNAKE_CASE__ ) for node in self] )
def lowerCAmelCase_ ( __A : SortedLinkedList , __A : SortedLinkedList ):
'''simple docstring'''
return SortedLinkedList(list(__A ) + list(__A ) )
if __name__ == "__main__":
import doctest
doctest.testmod()
__UpperCAmelCase = SortedLinkedList
print(merge_lists(SSL(test_data_odd), SSL(test_data_even))) | 692 |
'''simple docstring'''
from typing import Any, Dict, Optional
import torch
import torch.nn.functional as F
from torch import nn
from ..utils import maybe_allow_in_graph
from .activations import get_activation
from .attention_processor import Attention
from .embeddings import CombinedTimestepLabelEmbeddings
@maybe_allow_in_graph
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "layer_norm" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = only_cross_attention
snake_case: Optional[Any] = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm_zero'
snake_case: Tuple = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm'
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
raise ValueError(
F"""`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"""
F""" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}.""" )
# Define 3 blocks. Each block has its own normalization layer.
# 1. Self-Attn
if self.use_ada_layer_norm:
snake_case: List[str] = AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case: str = AdaLayerNormZero(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
else:
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if only_cross_attention else None , upcast_attention=SCREAMING_SNAKE_CASE__ , )
# 2. Cross-Attn
if cross_attention_dim is not None or double_self_attention:
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
# the second cross attention block.
snake_case: Tuple = (
AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm
else nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if not double_self_attention else None , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , upcast_attention=SCREAMING_SNAKE_CASE__ , ) # is self-attn if encoder_hidden_states is none
else:
snake_case: int = None
snake_case: Tuple = None
# 3. Feed-forward
snake_case: Union[str, Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = FeedForward(SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , activation_fn=SCREAMING_SNAKE_CASE__ , final_dropout=SCREAMING_SNAKE_CASE__ )
# let chunk size default to None
snake_case: Any = None
snake_case: Any = 0
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = chunk_size
snake_case: str = dim
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
if self.use_ada_layer_norm:
snake_case: Optional[int] = self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case , snake_case , snake_case , snake_case , snake_case: int = self.norma(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=hidden_states.dtype )
else:
snake_case: List[str] = self.norma(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = cross_attention_kwargs if cross_attention_kwargs is not None else {}
snake_case: List[str] = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=encoder_hidden_states if self.only_cross_attention else None , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if self.use_ada_layer_norm_zero:
snake_case: Tuple = gate_msa.unsqueeze(1 ) * attn_output
snake_case: List[str] = attn_output + hidden_states
# 2. Cross-Attention
if self.attna is not None:
snake_case: Dict = (
self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) if self.use_ada_layer_norm else self.norma(SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: List[str] = attn_output + hidden_states
# 3. Feed-forward
snake_case: str = self.norma(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: str = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
if self._chunk_size is not None:
# "feed_forward_chunk_size" can be used to save memory
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
raise ValueError(
F"""`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`.""" )
snake_case: List[str] = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
snake_case: Optional[Any] = torch.cat(
[self.ff(SCREAMING_SNAKE_CASE__ ) for hid_slice in norm_hidden_states.chunk(SCREAMING_SNAKE_CASE__ , dim=self._chunk_dim )] , dim=self._chunk_dim , )
else:
snake_case: int = self.ff(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: Union[str, Any] = gate_mlp.unsqueeze(1 ) * ff_output
snake_case: Tuple = ff_output + hidden_states
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 4 , SCREAMING_SNAKE_CASE__ = 0.0 , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: int = int(dim * mult )
snake_case: Optional[Any] = dim_out if dim_out is not None else dim
if activation_fn == "gelu":
snake_case: int = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if activation_fn == "gelu-approximate":
snake_case: Optional[Any] = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , approximate='tanh' )
elif activation_fn == "geglu":
snake_case: List[Any] = GEGLU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif activation_fn == "geglu-approximate":
snake_case: Optional[int] = ApproximateGELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.ModuleList([] )
# project in
self.net.append(SCREAMING_SNAKE_CASE__ )
# project dropout
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
# project out
self.net.append(nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
if final_dropout:
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for module in self.net:
snake_case: Optional[int] = module(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "none" ):
'''simple docstring'''
super().__init__()
snake_case: Optional[int] = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = approximate
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ , approximate=self.approximate )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) , approximate=self.approximate ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.proj(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.gelu(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = nn.Linear(SCREAMING_SNAKE_CASE__ , dim_out * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: int = self.proj(SCREAMING_SNAKE_CASE__ ).chunk(2 , dim=-1 )
return hidden_states * self.gelu(SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = self.proj(SCREAMING_SNAKE_CASE__ )
return x * torch.sigmoid(1.7_02 * x )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Optional[Any] = nn.Embedding(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = nn.SiLU()
snake_case: Union[str, Any] = nn.Linear(SCREAMING_SNAKE_CASE__ , embedding_dim * 2 )
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case: Dict = torch.chunk(SCREAMING_SNAKE_CASE__ , 2 )
snake_case: str = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale) + shift
return x
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = CombinedTimestepLabelEmbeddings(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.SiLU()
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , 6 * embedding_dim , bias=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ , eps=1E-6 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
snake_case: int = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case , snake_case , snake_case , snake_case , snake_case: str = emb.chunk(6 , dim=1 )
snake_case: Dict = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale_msa[:, None]) + shift_msa[:, None]
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 1E-5 ):
'''simple docstring'''
super().__init__()
snake_case: str = num_groups
snake_case: str = eps
if act_fn is None:
snake_case: Dict = None
else:
snake_case: List[str] = get_activation(SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , out_dim * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.act:
snake_case: Optional[Any] = self.act(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.linear(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = emb[:, :, None, None]
snake_case , snake_case: List[Any] = emb.chunk(2 , dim=1 )
snake_case: Any = F.group_norm(SCREAMING_SNAKE_CASE__ , self.num_groups , eps=self.eps )
snake_case: Optional[int] = x * (1 + scale) + shift
return x | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: str = [0] * len(__A )
snake_case: Tuple = []
snake_case: Tuple = [1] * len(__A )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(__A ) ):
if indegree[i] == 0:
queue.append(__A )
while queue:
snake_case: int = queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
snake_case: Any = long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(__A )
print(max(__A ) )
# Adjacency list of Graph
__UpperCAmelCase = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph) | 692 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
EulerAncestralDiscreteScheduler,
LMSDiscreteScheduler,
PNDMScheduler,
StableDiffusionPanoramaPipeline,
UNetaDConditionModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu, skip_mps
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
@skip_mps
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = StableDiffusionPanoramaPipeline
__UpperCamelCase = TEXT_TO_IMAGE_PARAMS
__UpperCamelCase = TEXT_TO_IMAGE_BATCH_PARAMS
__UpperCamelCase = TEXT_TO_IMAGE_IMAGE_PARAMS
__UpperCamelCase = TEXT_TO_IMAGE_IMAGE_PARAMS
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Optional[Any] = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=1 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('DownBlock2D', 'CrossAttnDownBlock2D') , up_block_types=('CrossAttnUpBlock2D', 'UpBlock2D') , cross_attention_dim=32 , )
snake_case: Dict = DDIMScheduler()
torch.manual_seed(0 )
snake_case: Tuple = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , )
torch.manual_seed(0 )
snake_case: Any = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1E-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=10_00 , )
snake_case: Optional[int] = CLIPTextModel(SCREAMING_SNAKE_CASE__ )
snake_case: str = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
snake_case: Optional[int] = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'safety_checker': None,
'feature_extractor': None,
}
return components
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0 ):
'''simple docstring'''
snake_case: Union[str, Any] = torch.manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {
'prompt': 'a photo of the dolomites',
'generator': generator,
# Setting height and width to None to prevent OOMs on CPU.
'height': None,
'width': None,
'num_inference_steps': 1,
'guidance_scale': 6.0,
'output_type': 'numpy',
}
return inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Optional[int] = self.get_dummy_components()
snake_case: Union[str, Any] = StableDiffusionPanoramaPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: Any = sd_pipe.to(SCREAMING_SNAKE_CASE__ )
sd_pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = sd_pipe(**SCREAMING_SNAKE_CASE__ ).images
snake_case: str = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: Optional[Any] = np.array([0.61_86, 0.53_74, 0.49_15, 0.41_35, 0.41_14, 0.45_63, 0.51_28, 0.49_77, 0.47_57] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
super().test_inference_batch_consistent(batch_sizes=[1, 2] )
def _UpperCamelCase ( self ):
'''simple docstring'''
super().test_inference_batch_single_identical(batch_size=2 , expected_max_diff=3.25E-3 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Union[str, Any] = self.get_dummy_components()
snake_case: int = StableDiffusionPanoramaPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: int = sd_pipe.to(SCREAMING_SNAKE_CASE__ )
sd_pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = 'french fries'
snake_case: List[str] = sd_pipe(**SCREAMING_SNAKE_CASE__ , negative_prompt=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = output.images
snake_case: Tuple = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: List[Any] = np.array([0.61_87, 0.53_75, 0.49_15, 0.41_36, 0.41_14, 0.45_63, 0.51_28, 0.49_76, 0.47_57] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Any = self.get_dummy_components()
snake_case: List[Any] = StableDiffusionPanoramaPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = sd_pipe.to(SCREAMING_SNAKE_CASE__ )
sd_pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = sd_pipe(**SCREAMING_SNAKE_CASE__ , view_batch_size=2 )
snake_case: List[str] = output.images
snake_case: str = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: List[str] = np.array([0.61_87, 0.53_75, 0.49_15, 0.41_36, 0.41_14, 0.45_63, 0.51_28, 0.49_76, 0.47_57] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: str = self.get_dummy_components()
snake_case: Any = EulerAncestralDiscreteScheduler(
beta_start=0.0_00_85 , beta_end=0.0_12 , beta_schedule='scaled_linear' )
snake_case: List[str] = StableDiffusionPanoramaPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = sd_pipe.to(SCREAMING_SNAKE_CASE__ )
sd_pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: int = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = sd_pipe(**SCREAMING_SNAKE_CASE__ ).images
snake_case: List[str] = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: Tuple = np.array([0.40_24, 0.65_10, 0.49_01, 0.53_78, 0.58_13, 0.56_22, 0.47_95, 0.44_67, 0.49_52] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Any = self.get_dummy_components()
snake_case: Dict = PNDMScheduler(
beta_start=0.0_00_85 , beta_end=0.0_12 , beta_schedule='scaled_linear' , skip_prk_steps=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = StableDiffusionPanoramaPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: int = sd_pipe.to(SCREAMING_SNAKE_CASE__ )
sd_pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: int = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = sd_pipe(**SCREAMING_SNAKE_CASE__ ).images
snake_case: Tuple = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: str = np.array([0.63_91, 0.62_91, 0.48_61, 0.51_34, 0.55_52, 0.45_78, 0.50_32, 0.50_23, 0.45_39] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
@slow
@require_torch_gpu
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__=0 ):
'''simple docstring'''
snake_case: Optional[Any] = torch.manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: str = {
'prompt': 'a photo of the dolomites',
'generator': generator,
'num_inference_steps': 3,
'guidance_scale': 7.5,
'output_type': 'numpy',
}
return inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = 'stabilityai/stable-diffusion-2-base'
snake_case: Optional[Any] = DDIMScheduler.from_pretrained(SCREAMING_SNAKE_CASE__ , subfolder='scheduler' )
snake_case: List[str] = StableDiffusionPanoramaPipeline.from_pretrained(SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ , safety_checker=SCREAMING_SNAKE_CASE__ )
pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
pipe.enable_attention_slicing()
snake_case: int = self.get_inputs()
snake_case: Dict = pipe(**SCREAMING_SNAKE_CASE__ ).images
snake_case: Tuple = image[0, -3:, -3:, -1].flatten()
assert image.shape == (1, 5_12, 20_48, 3)
snake_case: List[str] = np.array(
[
0.36_96_83_92,
0.27_02_53_72,
0.32_44_67_66,
0.28_37_93_87,
0.36_36_32_74,
0.30_73_33_47,
0.27_10_00_27,
0.27_05_41_25,
0.25_53_60_96,
] )
assert np.abs(expected_slice - image_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = StableDiffusionPanoramaPipeline.from_pretrained(
'stabilityai/stable-diffusion-2-base' , safety_checker=SCREAMING_SNAKE_CASE__ )
snake_case: Any = LMSDiscreteScheduler.from_config(pipe.scheduler.config )
pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
pipe.enable_attention_slicing()
snake_case: Tuple = self.get_inputs()
snake_case: Union[str, Any] = pipe(**SCREAMING_SNAKE_CASE__ ).images
snake_case: Optional[int] = image[0, -3:, -3:, -1].flatten()
assert image.shape == (1, 5_12, 20_48, 3)
snake_case: str = np.array(
[
[
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
]
] )
assert np.abs(expected_slice - image_slice ).max() < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = 0
def callback_fn(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) -> None:
snake_case: Any = True
nonlocal number_of_steps
number_of_steps += 1
if step == 1:
snake_case: Tuple = latents.detach().cpu().numpy()
assert latents.shape == (1, 4, 64, 2_56)
snake_case: str = latents[0, -3:, -3:, -1]
snake_case: Optional[int] = np.array(
[
0.18_68_18_69,
0.33_90_78_16,
0.5_36_12_76,
0.14_43_28_65,
-0.02_85_66_11,
-0.73_94_11_23,
0.23_39_79_87,
0.47_32_26_82,
-0.37_82_31_64,
] )
assert np.abs(latents_slice.flatten() - expected_slice ).max() < 5E-2
elif step == 2:
snake_case: str = latents.detach().cpu().numpy()
assert latents.shape == (1, 4, 64, 2_56)
snake_case: List[Any] = latents[0, -3:, -3:, -1]
snake_case: Any = np.array(
[
0.18_53_96_45,
0.33_98_72_48,
0.5_37_85_59,
0.14_43_71_42,
-0.02_45_52_61,
-0.7_33_83_17,
0.23_99_07_55,
0.47_35_62_72,
-0.3_78_65_05,
] )
assert np.abs(latents_slice.flatten() - expected_slice ).max() < 5E-2
snake_case: Any = False
snake_case: List[str] = 'stabilityai/stable-diffusion-2-base'
snake_case: Dict = DDIMScheduler.from_pretrained(SCREAMING_SNAKE_CASE__ , subfolder='scheduler' )
snake_case: Union[str, Any] = StableDiffusionPanoramaPipeline.from_pretrained(SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ , safety_checker=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
pipe.enable_attention_slicing()
snake_case: int = self.get_inputs()
pipe(**SCREAMING_SNAKE_CASE__ , callback=SCREAMING_SNAKE_CASE__ , callback_steps=1 )
assert callback_fn.has_been_called
assert number_of_steps == 3
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.cuda.empty_cache()
torch.cuda.reset_max_memory_allocated()
torch.cuda.reset_peak_memory_stats()
snake_case: Tuple = 'stabilityai/stable-diffusion-2-base'
snake_case: Union[str, Any] = DDIMScheduler.from_pretrained(SCREAMING_SNAKE_CASE__ , subfolder='scheduler' )
snake_case: str = StableDiffusionPanoramaPipeline.from_pretrained(SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ , safety_checker=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
pipe.enable_attention_slicing(1 )
pipe.enable_sequential_cpu_offload()
snake_case: Tuple = self.get_inputs()
snake_case: Optional[int] = pipe(**SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = torch.cuda.max_memory_allocated()
# make sure that less than 5.2 GB is allocated
assert mem_bytes < 5.5 * 10**9 | 692 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from . import (
albert,
align,
altclip,
audio_spectrogram_transformer,
auto,
autoformer,
bark,
bart,
barthez,
bartpho,
beit,
bert,
bert_generation,
bert_japanese,
bertweet,
big_bird,
bigbird_pegasus,
biogpt,
bit,
blenderbot,
blenderbot_small,
blip,
blip_a,
bloom,
bridgetower,
byta,
camembert,
canine,
chinese_clip,
clap,
clip,
clipseg,
codegen,
conditional_detr,
convbert,
convnext,
convnextva,
cpm,
cpmant,
ctrl,
cvt,
dataavec,
deberta,
deberta_va,
decision_transformer,
deformable_detr,
deit,
deprecated,
deta,
detr,
dialogpt,
dinat,
distilbert,
dit,
donut,
dpr,
dpt,
efficientformer,
efficientnet,
electra,
encodec,
encoder_decoder,
ernie,
ernie_m,
esm,
falcon,
flaubert,
flava,
fnet,
focalnet,
fsmt,
funnel,
git,
glpn,
gpta,
gpt_bigcode,
gpt_neo,
gpt_neox,
gpt_neox_japanese,
gpt_swa,
gptj,
gptsan_japanese,
graphormer,
groupvit,
herbert,
hubert,
ibert,
imagegpt,
informer,
instructblip,
jukebox,
layoutlm,
layoutlmva,
layoutlmva,
layoutxlm,
led,
levit,
lilt,
llama,
longformer,
longta,
luke,
lxmert,
mam_aaa,
marian,
markuplm,
maskaformer,
maskformer,
mbart,
mbartaa,
mega,
megatron_bert,
megatron_gpta,
mgp_str,
mluke,
mobilebert,
mobilenet_va,
mobilenet_va,
mobilevit,
mobilevitva,
mpnet,
mra,
mta,
musicgen,
mvp,
nat,
nezha,
nllb,
nllb_moe,
nystromformer,
oneformer,
open_llama,
openai,
opt,
owlvit,
pegasus,
pegasus_x,
perceiver,
phobert,
pixastruct,
plbart,
poolformer,
prophetnet,
qdqbert,
rag,
realm,
reformer,
regnet,
rembert,
resnet,
roberta,
roberta_prelayernorm,
roc_bert,
roformer,
rwkv,
sam,
segformer,
sew,
sew_d,
speech_encoder_decoder,
speech_to_text,
speech_to_text_a,
speechta,
splinter,
squeezebert,
swiftformer,
swin,
swinasr,
swinva,
switch_transformers,
ta,
table_transformer,
tapas,
time_series_transformer,
timesformer,
timm_backbone,
transfo_xl,
trocr,
tvlt,
umta,
unispeech,
unispeech_sat,
upernet,
videomae,
vilt,
vision_encoder_decoder,
vision_text_dual_encoder,
visual_bert,
vit,
vit_hybrid,
vit_mae,
vit_msn,
vivit,
wavaveca,
wavaveca_conformer,
wavaveca_phoneme,
wavaveca_with_lm,
wavlm,
whisper,
x_clip,
xglm,
xlm,
xlm_prophetnet,
xlm_roberta,
xlm_roberta_xl,
xlnet,
xmod,
yolos,
yoso,
) | 692 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_roformer": ["ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoFormerConfig", "RoFormerOnnxConfig"],
"tokenization_roformer": ["RoFormerTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["RoFormerTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoFormerForCausalLM",
"RoFormerForMaskedLM",
"RoFormerForMultipleChoice",
"RoFormerForQuestionAnswering",
"RoFormerForSequenceClassification",
"RoFormerForTokenClassification",
"RoFormerLayer",
"RoFormerModel",
"RoFormerPreTrainedModel",
"load_tf_weights_in_roformer",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFRoFormerForCausalLM",
"TFRoFormerForMaskedLM",
"TFRoFormerForMultipleChoice",
"TFRoFormerForQuestionAnswering",
"TFRoFormerForSequenceClassification",
"TFRoFormerForTokenClassification",
"TFRoFormerLayer",
"TFRoFormerModel",
"TFRoFormerPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"FlaxRoFormerForMaskedLM",
"FlaxRoFormerForMultipleChoice",
"FlaxRoFormerForQuestionAnswering",
"FlaxRoFormerForSequenceClassification",
"FlaxRoFormerForTokenClassification",
"FlaxRoFormerModel",
"FlaxRoFormerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig
from .tokenization_roformer import RoFormerTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_roformer_fast import RoFormerTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roformer import (
ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
RoFormerForCausalLM,
RoFormerForMaskedLM,
RoFormerForMultipleChoice,
RoFormerForQuestionAnswering,
RoFormerForSequenceClassification,
RoFormerForTokenClassification,
RoFormerLayer,
RoFormerModel,
RoFormerPreTrainedModel,
load_tf_weights_in_roformer,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_roformer import (
TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TFRoFormerForCausalLM,
TFRoFormerForMaskedLM,
TFRoFormerForMultipleChoice,
TFRoFormerForQuestionAnswering,
TFRoFormerForSequenceClassification,
TFRoFormerForTokenClassification,
TFRoFormerLayer,
TFRoFormerModel,
TFRoFormerPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_roformer import (
FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
FlaxRoFormerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = "\n# Transformers 설치 방법\n! pip install transformers datasets\n# 마지막 릴리스 대신 소스에서 설치하려면, 위 명령을 주석으로 바꾸고 아래 명령을 해제하세요.\n# ! pip install git+https://github.com/huggingface/transformers.git\n"
__UpperCAmelCase = [{"type": "code", "content": INSTALL_CONTENT}]
__UpperCAmelCase = {
"{processor_class}": "FakeProcessorClass",
"{model_class}": "FakeModelClass",
"{object_class}": "FakeObjectClass",
} | 692 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 | 1 |
'''simple docstring'''
import sys
import webbrowser
import requests
from bsa import BeautifulSoup
from fake_useragent import UserAgent
if __name__ == "__main__":
print("Googling.....")
__UpperCAmelCase = "https://www.google.com/search?q=" + " ".join(sys.argv[1:])
__UpperCAmelCase = requests.get(url, headers={"UserAgent": UserAgent().random})
# res.raise_for_status()
with open("project1a.html", "wb") as out_file: # only for knowing the class
for data in res.iter_content(10_000):
out_file.write(data)
__UpperCAmelCase = BeautifulSoup(res.text, "html.parser")
__UpperCAmelCase = list(soup.select(".eZt8xd"))[:5]
print(len(links))
for link in links:
if link.text == "Maps":
webbrowser.open(link.get("href"))
else:
webbrowser.open(F'https://google.com{link.get("href")}') | 692 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 | 1 |
'''simple docstring'''
from dataclasses import dataclass, field
from typing import Tuple
from ..utils import cached_property, is_tf_available, logging, requires_backends
from .benchmark_args_utils import BenchmarkArguments
if is_tf_available():
import tensorflow as tf
__UpperCAmelCase = logging.get_logger(__name__)
@dataclass
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = [
"no_inference",
"no_cuda",
"no_tpu",
"no_speed",
"no_memory",
"no_env_print",
"no_multi_process",
]
def __init__( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for deprecated_arg in self.deprecated_args:
if deprecated_arg in kwargs:
snake_case: Union[str, Any] = deprecated_arg[3:]
snake_case: Any = not kwargs.pop(SCREAMING_SNAKE_CASE__ )
logger.warning(
F"""{deprecated_arg} is depreciated. Please use --no-{positive_arg} or"""
F""" {positive_arg}={kwargs[positive_arg]}""" )
snake_case: int = kwargs.pop('tpu_name' , self.tpu_name )
snake_case: Tuple = kwargs.pop('device_idx' , self.device_idx )
snake_case: str = kwargs.pop('eager_mode' , self.eager_mode )
snake_case: str = kwargs.pop('use_xla' , self.use_xla )
super().__init__(**SCREAMING_SNAKE_CASE__ )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Name of TPU"} , )
__UpperCamelCase = field(
default=0 , metadata={"help": "CPU / GPU device index. Defaults to 0."} , )
__UpperCamelCase = field(default=snake_case , metadata={"help": "Benchmark models in eager model."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": "Benchmark models using XLA JIT compilation. Note that `eager_model` has to be set to `False`."
} , )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
snake_case: List[Any] = None
if self.tpu:
try:
if self.tpu_name:
snake_case: List[Any] = tf.distribute.cluster_resolver.TPUClusterResolver(self.tpu_name )
else:
snake_case: Dict = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
snake_case: Any = None
return tpu
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
if self.is_tpu:
tf.config.experimental_connect_to_cluster(self._setup_tpu )
tf.tpu.experimental.initialize_tpu_system(self._setup_tpu )
snake_case: Union[str, Any] = tf.distribute.TPUStrategy(self._setup_tpu )
else:
# currently no multi gpu is allowed
if self.is_gpu:
# TODO: Currently only single GPU is supported
tf.config.set_visible_devices(self.gpu_list[self.device_idx] , 'GPU' )
snake_case: str = tf.distribute.OneDeviceStrategy(device=F"""/gpu:{self.device_idx}""" )
else:
tf.config.set_visible_devices([] , 'GPU' ) # disable GPU
snake_case: str = tf.distribute.OneDeviceStrategy(device=F"""/cpu:{self.device_idx}""" )
return strategy
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
return self._setup_tpu is not None
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
return self._setup_strategy
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
return tf.config.list_physical_devices('GPU' )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
requires_backends(self , ['tf'] )
if self.cuda:
return len(self.gpu_list )
return 0
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.n_gpu > 0 | 692 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int , __A : Any ):
'''simple docstring'''
snake_case: int = 0
while b > 0:
if b & 1:
res += a
a += a
b >>= 1
return res
def lowerCAmelCase_ ( __A : int , __A : Optional[int] , __A : List[Any] ):
'''simple docstring'''
snake_case: List[Any] = 0
while b > 0:
if b & 1:
snake_case: Union[str, Any] = ((res % c) + (a % c)) % c
a += a
b >>= 1
return res | 692 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel
from diffusers import DDIMScheduler, LDMPipeline, UNetaDModel, VQModel
from diffusers.utils.testing_utils import enable_full_determinism, require_torch, slow, torch_device
enable_full_determinism()
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Union[str, Any] = UNetaDModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=3 , out_channels=3 , down_block_types=('DownBlock2D', 'AttnDownBlock2D') , up_block_types=('AttnUpBlock2D', 'UpBlock2D') , )
return model
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: str = VQModel(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=3 , )
return model
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Tuple = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1E-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=10_00 , )
return CLIPTextModel(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.dummy_uncond_unet
snake_case: Optional[Any] = DDIMScheduler()
snake_case: Union[str, Any] = self.dummy_vq_model
snake_case: Union[str, Any] = LDMPipeline(unet=SCREAMING_SNAKE_CASE__ , vqvae=SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ )
ldm.to(SCREAMING_SNAKE_CASE__ )
ldm.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = torch.manual_seed(0 )
snake_case: Dict = ldm(generator=SCREAMING_SNAKE_CASE__ , num_inference_steps=2 , output_type='numpy' ).images
snake_case: Union[str, Any] = torch.manual_seed(0 )
snake_case: List[Any] = ldm(generator=SCREAMING_SNAKE_CASE__ , num_inference_steps=2 , output_type='numpy' , return_dict=SCREAMING_SNAKE_CASE__ )[0]
snake_case: Union[str, Any] = image[0, -3:, -3:, -1]
snake_case: Tuple = image_from_tuple[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
snake_case: Optional[int] = np.array([0.85_12, 0.8_18, 0.64_11, 0.68_08, 0.44_65, 0.56_18, 0.46, 0.62_31, 0.51_72] )
snake_case: Union[str, Any] = 1E-2 if torch_device != 'mps' else 3E-2
assert np.abs(image_slice.flatten() - expected_slice ).max() < tolerance
assert np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() < tolerance
@slow
@require_torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = LDMPipeline.from_pretrained('CompVis/ldm-celebahq-256' )
ldm.to(SCREAMING_SNAKE_CASE__ )
ldm.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: int = torch.manual_seed(0 )
snake_case: Union[str, Any] = ldm(generator=SCREAMING_SNAKE_CASE__ , num_inference_steps=5 , output_type='numpy' ).images
snake_case: Any = image[0, -3:, -3:, -1]
assert image.shape == (1, 2_56, 2_56, 3)
snake_case: Optional[int] = np.array([0.43_99, 0.4_49_75, 0.4_68_25, 0.4_74, 0.43_59, 0.45_81, 0.4_50_95, 0.43_41, 0.44_47] )
snake_case: List[Any] = 1E-2 if torch_device != 'mps' else 3E-2
assert np.abs(image_slice.flatten() - expected_slice ).max() < tolerance | 692 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = "▁"
__UpperCAmelCase = {"vocab_file": "sentencepiece.bpe.model"}
__UpperCAmelCase = {
"vocab_file": {
"facebook/xglm-564M": "https://huggingface.co/facebook/xglm-564M/resolve/main/sentencepiece.bpe.model",
}
}
__UpperCAmelCase = {
"facebook/xglm-564M": 2_048,
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
# Compatibility with the original tokenizer
snake_case: Optional[Any] = 7
snake_case: List[str] = [F"""<madeupword{i}>""" for i in range(self.num_madeup_words )]
snake_case: str = kwargs.get('additional_special_tokens' , [] )
kwargs["additional_special_tokens"] += [
word for word in madeup_words if word not in kwargs["additional_special_tokens"]
]
super().__init__(
bos_token=SCREAMING_SNAKE_CASE__ , eos_token=SCREAMING_SNAKE_CASE__ , unk_token=SCREAMING_SNAKE_CASE__ , sep_token=SCREAMING_SNAKE_CASE__ , cls_token=SCREAMING_SNAKE_CASE__ , pad_token=SCREAMING_SNAKE_CASE__ , sp_model_kwargs=self.sp_model_kwargs , **SCREAMING_SNAKE_CASE__ , )
snake_case: int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(SCREAMING_SNAKE_CASE__ ) )
snake_case: int = vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
snake_case: Tuple = 1
# Mimic fairseq token-to-id alignment for the first 4 token
snake_case: Optional[Any] = {'<s>': 0, '<pad>': 1, '</s>': 2, '<unk>': 3}
snake_case: Union[str, Any] = len(self.sp_model )
snake_case: str = {F"""<madeupword{i}>""": sp_size + i + self.fairseq_offset for i in range(self.num_madeup_words )}
self.fairseq_tokens_to_ids.update(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def __getstate__( self ):
'''simple docstring'''
snake_case: List[Any] = self.__dict__.copy()
snake_case: Union[str, Any] = None
snake_case: Union[str, Any] = self.sp_model.serialized_model_proto()
return state
def __setstate__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
snake_case: Union[str, Any] = {}
snake_case: Tuple = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.LoadFromSerializedProto(self.sp_model_proto )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.sep_token_id] + token_ids_a
snake_case: Optional[Any] = [self.sep_token_id]
return sep + token_ids_a + sep + sep + token_ids_a
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=SCREAMING_SNAKE_CASE__ , token_ids_a=SCREAMING_SNAKE_CASE__ , already_has_special_tokens=SCREAMING_SNAKE_CASE__ )
if token_ids_a is None:
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ )) + [1, 1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
if token_ids_a is None:
return len(sep + token_ids_a ) * [0]
return len(sep + token_ids_a + sep + sep + token_ids_a ) * [0]
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.sp_model ) + self.fairseq_offset + self.num_madeup_words
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = {self.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.sp_model.encode(SCREAMING_SNAKE_CASE__ , out_type=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
snake_case: Dict = self.sp_model.PieceToId(SCREAMING_SNAKE_CASE__ )
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = ''.join(SCREAMING_SNAKE_CASE__ ).replace(SCREAMING_SNAKE_CASE__ , ' ' ).strip()
return out_string
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(SCREAMING_SNAKE_CASE__ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: List[str] = os.path.join(
SCREAMING_SNAKE_CASE__ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(SCREAMING_SNAKE_CASE__ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , SCREAMING_SNAKE_CASE__ )
elif not os.path.isfile(self.vocab_file ):
with open(SCREAMING_SNAKE_CASE__ , 'wb' ) as fi:
snake_case: int = self.sp_model.serialized_model_proto()
fi.write(SCREAMING_SNAKE_CASE__ )
return (out_vocab_file,) | 692 | 1 |
'''simple docstring'''
import argparse
from pathlib import Path
from typing import Dict, OrderedDict, Tuple
import torch
from audiocraft.models import MusicGen
from transformers import (
AutoFeatureExtractor,
AutoTokenizer,
EncodecModel,
MusicgenDecoderConfig,
MusicgenForConditionalGeneration,
MusicgenProcessor,
TaEncoderModel,
)
from transformers.models.musicgen.modeling_musicgen import MusicgenForCausalLM
from transformers.utils import logging
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = ["model.decoder.embed_positions.weights"]
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "emb" in name:
snake_case: Optional[int] = name.replace('emb' , 'model.decoder.embed_tokens' )
if "transformer" in name:
snake_case: Any = name.replace('transformer' , 'model.decoder' )
if "cross_attention" in name:
snake_case: Optional[Any] = name.replace('cross_attention' , 'encoder_attn' )
if "linear1" in name:
snake_case: Tuple = name.replace('linear1' , 'fc1' )
if "linear2" in name:
snake_case: int = name.replace('linear2' , 'fc2' )
if "norm1" in name:
snake_case: List[Any] = name.replace('norm1' , 'self_attn_layer_norm' )
if "norm_cross" in name:
snake_case: List[Any] = name.replace('norm_cross' , 'encoder_attn_layer_norm' )
if "norm2" in name:
snake_case: int = name.replace('norm2' , 'final_layer_norm' )
if "out_norm" in name:
snake_case: Optional[int] = name.replace('out_norm' , 'model.decoder.layer_norm' )
if "linears" in name:
snake_case: int = name.replace('linears' , 'lm_heads' )
if "condition_provider.conditioners.description.output_proj" in name:
snake_case: Optional[int] = name.replace('condition_provider.conditioners.description.output_proj' , 'enc_to_dec_proj' )
return name
def lowerCAmelCase_ ( __A : OrderedDict , __A : int ):
'''simple docstring'''
snake_case: Tuple = list(state_dict.keys() )
snake_case: Union[str, Any] = {}
for key in keys:
snake_case: Tuple = state_dict.pop(__A )
snake_case: Union[str, Any] = rename_keys(__A )
if "in_proj_weight" in key:
# split fused qkv proj
snake_case: str = val[:hidden_size, :]
snake_case: Dict = val[hidden_size : 2 * hidden_size, :]
snake_case: Any = val[-hidden_size:, :]
elif "enc_to_dec_proj" in key:
snake_case: int = val
else:
snake_case: Any = val
return state_dict, enc_dec_proj_state_dict
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
if checkpoint == "small":
# default config values
snake_case: Dict = 10_24
snake_case: Tuple = 24
snake_case: str = 16
elif checkpoint == "medium":
snake_case: str = 15_36
snake_case: Dict = 48
snake_case: Optional[Any] = 24
elif checkpoint == "large":
snake_case: List[str] = 20_48
snake_case: int = 48
snake_case: List[str] = 32
else:
raise ValueError(f"""Checkpoint should be one of `['small', 'medium', 'large']`, got {checkpoint}.""" )
snake_case: Union[str, Any] = MusicgenDecoderConfig(
hidden_size=__A , ffn_dim=hidden_size * 4 , num_hidden_layers=__A , num_attention_heads=__A , )
return config
@torch.no_grad()
def lowerCAmelCase_ ( __A : str , __A : List[Any]=None , __A : Dict=None , __A : List[Any]="cpu" ):
'''simple docstring'''
snake_case: int = MusicGen.get_pretrained(__A , device=__A )
snake_case: Dict = decoder_config_from_checkpoint(__A )
snake_case: List[str] = fairseq_model.lm.state_dict()
snake_case , snake_case: List[Any] = rename_state_dict(
__A , hidden_size=decoder_config.hidden_size )
snake_case: List[Any] = TaEncoderModel.from_pretrained('t5-base' )
snake_case: List[Any] = EncodecModel.from_pretrained('facebook/encodec_32khz' )
snake_case: Tuple = MusicgenForCausalLM(__A ).eval()
# load all decoder weights - expect that we'll be missing embeddings and enc-dec projection
snake_case , snake_case: List[Any] = decoder.load_state_dict(__A , strict=__A )
for key in missing_keys.copy():
if key.startswith(('text_encoder', 'audio_encoder') ) or key in EXPECTED_MISSING_KEYS:
missing_keys.remove(__A )
if len(__A ) > 0:
raise ValueError(f"""Missing key(s) in state_dict: {missing_keys}""" )
if len(__A ) > 0:
raise ValueError(f"""Unexpected key(s) in state_dict: {unexpected_keys}""" )
# init the composite model
snake_case: Optional[Any] = MusicgenForConditionalGeneration(text_encoder=__A , audio_encoder=__A , decoder=__A )
# load the pre-trained enc-dec projection (from the decoder state dict)
model.enc_to_dec_proj.load_state_dict(__A )
# check we can do a forward pass
snake_case: Optional[int] = torch.arange(0 , 8 , dtype=torch.long ).reshape(2 , -1 )
snake_case: List[str] = input_ids.reshape(2 * 4 , -1 )
with torch.no_grad():
snake_case: Union[str, Any] = model(input_ids=__A , decoder_input_ids=__A ).logits
if logits.shape != (8, 1, 20_48):
raise ValueError('Incorrect shape for logits' )
# now construct the processor
snake_case: Union[str, Any] = AutoTokenizer.from_pretrained('t5-base' )
snake_case: Optional[Any] = AutoFeatureExtractor.from_pretrained('facebook/encodec_32khz' , padding_side='left' )
snake_case: Optional[Any] = MusicgenProcessor(feature_extractor=__A , tokenizer=__A )
# set the appropriate bos/pad token ids
snake_case: str = 20_48
snake_case: Dict = 20_48
# set other default generation config params
snake_case: str = int(30 * audio_encoder.config.frame_rate )
snake_case: Union[str, Any] = True
snake_case: Union[str, Any] = 3.0
if pytorch_dump_folder is not None:
Path(__A ).mkdir(exist_ok=__A )
logger.info(f"""Saving model {checkpoint} to {pytorch_dump_folder}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if repo_id:
logger.info(f"""Pushing model {checkpoint} to {repo_id}""" )
model.push_to_hub(__A )
processor.push_to_hub(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--checkpoint",
default="small",
type=str,
help="Checkpoint size of the MusicGen model you'd like to convert. Can be one of: `['small', 'medium', 'large']`.",
)
parser.add_argument(
"--pytorch_dump_folder",
required=True,
default=None,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub", default=None, type=str, help="Where to upload the converted model on the 🤗 hub."
)
parser.add_argument(
"--device", default="cpu", type=str, help="Torch device to run the conversion, either cpu or cuda."
)
__UpperCAmelCase = parser.parse_args()
convert_musicgen_checkpoint(args.checkpoint, args.pytorch_dump_folder, args.push_to_hub) | 692 |
'''simple docstring'''
from operator import delitem, getitem, setitem
import pytest
from data_structures.hashing.hash_map import HashMap
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
return getitem, k
def lowerCAmelCase_ ( __A : Any , __A : Optional[int] ):
'''simple docstring'''
return setitem, k, v
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
return delitem, k
def lowerCAmelCase_ ( __A : str , __A : int , *__A : Tuple ):
'''simple docstring'''
try:
return fun(__A , *__A ), None
except Exception as e:
return None, e
__UpperCAmelCase = (
_set("key_a", "val_a"),
_set("key_b", "val_b"),
)
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_a", "val_b"),
]
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_b", "val_b"),
_del("key_a"),
_del("key_b"),
_set("key_a", "val_a"),
_del("key_a"),
]
__UpperCAmelCase = [
_get("key_a"),
_del("key_a"),
_set("key_a", "val_a"),
_del("key_a"),
_del("key_a"),
_get("key_a"),
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
*[_del(x) for x in range(5)],
_set("key_a", "val_b"),
]
@pytest.mark.parametrize(
'operations' , (
pytest.param(_add_items , id='add items' ),
pytest.param(_overwrite_items , id='overwrite items' ),
pytest.param(_delete_items , id='delete items' ),
pytest.param(_access_absent_items , id='access absent items' ),
pytest.param(_add_with_resize_up , id='add with resize up' ),
pytest.param(_add_with_resize_down , id='add with resize down' ),
) , )
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[Any] = HashMap(initial_block_size=4 )
snake_case: List[Any] = {}
for _, (fun, *args) in enumerate(__A ):
snake_case , snake_case: Optional[int] = _run_operation(__A , __A , *__A )
snake_case , snake_case: str = _run_operation(__A , __A , *__A )
assert my_res == py_res
assert str(__A ) == str(__A )
assert set(__A ) == set(__A )
assert len(__A ) == len(__A )
assert set(my.items() ) == set(py.items() )
def lowerCAmelCase_ ( ):
'''simple docstring'''
def is_public(__A : str ) -> bool:
return not name.startswith('_' )
snake_case: Dict = {name for name in dir({} ) if is_public(__A )}
snake_case: List[str] = {name for name in dir(HashMap() ) if is_public(__A )}
assert dict_public_names > hash_public_names | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
import random
import unittest
from transformers import TransfoXLConfig, is_tf_available
from transformers.testing_utils import require_tf, slow
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import (
TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST,
TFTransfoXLForSequenceClassification,
TFTransfoXLLMHeadModel,
TFTransfoXLModel,
)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: str = parent
snake_case: Tuple = 13
snake_case: List[Any] = 7
snake_case: Tuple = 30
snake_case: Dict = self.seq_length + self.mem_len
snake_case: Any = 15
snake_case: Optional[int] = True
snake_case: Any = True
snake_case: List[Any] = 99
snake_case: Union[str, Any] = [10, 50, 80]
snake_case: Any = 32
snake_case: int = 32
snake_case: str = 4
snake_case: List[Any] = 8
snake_case: Any = 1_28
snake_case: Optional[Any] = 2
snake_case: Union[str, Any] = 2
snake_case: Dict = None
snake_case: List[str] = 1
snake_case: List[Any] = 0
snake_case: Dict = 3
snake_case: Optional[Any] = self.vocab_size - 1
snake_case: Optional[int] = 0.01
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: str = None
if self.use_labels:
snake_case: int = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: int = TransfoXLConfig(
vocab_size=self.vocab_size , mem_len=self.mem_len , clamp_len=self.clamp_len , cutoffs=self.cutoffs , d_model=self.hidden_size , d_embed=self.d_embed , n_head=self.num_attention_heads , d_head=self.d_head , d_inner=self.d_inner , div_val=self.div_val , n_layer=self.num_hidden_layers , eos_token_id=self.eos_token_id , pad_token_id=self.vocab_size - 1 , init_range=self.init_range , num_labels=self.num_labels , )
return (config, input_ids_a, input_ids_a, lm_labels)
def _UpperCamelCase ( self ):
'''simple docstring'''
random.seed(self.seed )
tf.random.set_seed(self.seed )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = TFTransfoXLModel(SCREAMING_SNAKE_CASE__ )
snake_case , snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ ).to_tuple()
snake_case: Dict = {'input_ids': input_ids_a, 'mems': mems_a}
snake_case , snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ ).to_tuple()
self.parent.assertEqual(hidden_states_a.shape , (self.batch_size, self.seq_length, self.hidden_size) )
self.parent.assertEqual(hidden_states_a.shape , (self.batch_size, self.seq_length, self.hidden_size) )
self.parent.assertListEqual(
[mem.shape for mem in mems_a] , [(self.mem_len, self.batch_size, self.hidden_size)] * self.num_hidden_layers , )
self.parent.assertListEqual(
[mem.shape for mem in mems_a] , [(self.mem_len, self.batch_size, self.hidden_size)] * self.num_hidden_layers , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = TFTransfoXLLMHeadModel(SCREAMING_SNAKE_CASE__ )
snake_case , snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ ).to_tuple()
snake_case: Dict = {'input_ids': input_ids_a, 'labels': lm_labels}
snake_case , snake_case: int = model(SCREAMING_SNAKE_CASE__ ).to_tuple()
snake_case , snake_case: Optional[Any] = model([input_ids_a, mems_a] ).to_tuple()
snake_case: Optional[Any] = {'input_ids': input_ids_a, 'mems': mems_a, 'labels': lm_labels}
snake_case , snake_case: str = model(SCREAMING_SNAKE_CASE__ ).to_tuple()
self.parent.assertEqual(lm_logits_a.shape , (self.batch_size, self.seq_length, self.vocab_size) )
self.parent.assertListEqual(
[mem.shape for mem in mems_a] , [(self.mem_len, self.batch_size, self.hidden_size)] * self.num_hidden_layers , )
self.parent.assertEqual(lm_logits_a.shape , (self.batch_size, self.seq_length, self.vocab_size) )
self.parent.assertListEqual(
[mem.shape for mem in mems_a] , [(self.mem_len, self.batch_size, self.hidden_size)] * self.num_hidden_layers , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = TFTransfoXLForSequenceClassification(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.prepare_config_and_inputs()
((snake_case) , (snake_case) , (snake_case) , (snake_case)): List[str] = config_and_inputs
snake_case: List[str] = {'input_ids': input_ids_a}
return config, inputs_dict
@require_tf
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (
(TFTransfoXLModel, TFTransfoXLLMHeadModel, TFTransfoXLForSequenceClassification) if is_tf_available() else ()
)
__UpperCamelCase = () if is_tf_available() else ()
__UpperCamelCase = (
{
"feature-extraction": TFTransfoXLModel,
"text-classification": TFTransfoXLForSequenceClassification,
"text-generation": TFTransfoXLLMHeadModel,
"zero-shot": TFTransfoXLForSequenceClassification,
}
if is_tf_available()
else {}
)
# TODO: add this test when TFTransfoXLLMHead has a linear output layer implemented
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if pipeline_test_casse_name == "TextGenerationPipelineTests":
# Get `ValueError: AttributeError: 'NoneType' object has no attribute 'new_ones'` or `AssertionError`.
# `TransfoXLConfig` was never used in pipeline tests: cannot create a simple
# tokenizer.
return True
return False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = TFTransfoXLModelTester(self )
snake_case: Dict = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , d_embed=37 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
def _UpperCamelCase ( self ):
'''simple docstring'''
self.model_tester.set_seed()
snake_case: Optional[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_transfo_xl_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.model_tester.set_seed()
snake_case: Optional[int] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_transfo_xl_lm_head(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_transfo_xl_for_sequence_classification(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: List[str] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[Any] = [TFTransfoXLForSequenceClassification]
for model_class in self.all_model_classes:
snake_case: Optional[int] = model_class(SCREAMING_SNAKE_CASE__ )
assert isinstance(model.get_input_embeddings() , tf.keras.layers.Layer )
if model_class in list_other_models_with_output_ebd:
snake_case: List[Any] = model.get_output_embeddings()
assert isinstance(SCREAMING_SNAKE_CASE__ , tf.keras.layers.Layer )
snake_case: Any = model.get_bias()
assert name is None
else:
snake_case: List[str] = model.get_output_embeddings()
assert x is None
snake_case: Optional[int] = model.get_bias()
assert name is None
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
for model_name in TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: str = TFTransfoXLModel.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
@unittest.skip(reason='This model doesn\'t play well with fit() due to not returning a single loss.' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@require_tf
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@unittest.skip('Skip test until #12651 is resolved.' )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = TFTransfoXLLMHeadModel.from_pretrained('transfo-xl-wt103' )
# fmt: off
snake_case: Dict = tf.convert_to_tensor([[33,12_97,2,1,10_09,4,11_09,1_17_39,47_62,3_58,5,25,2_45,22,17_06,17,2_00_98,5,32_15,21,37,11_10,3,13,10_41,4,24,6_03,4_90,2,7_14_77,2_00_98,10_44_47,2,2_09_61,1,26_04,4,1,3_29,3,62_24,8_31,1_60_02,2,8,6_03,7_89_67,2_95_46,23,8_03,20,25,4_16,5,8,2_32,4,2_77,6,18_55,46_01,3,2_95_46,54,8,36_09,5,5_72_11,49,4,1,2_77,18,8,17_55,1_56_91,3,3_41,25,4_16,6_93,4_25_73,71,17,4_01,94,31,1_79_19,2,2_95_46,78_73,18,1,4_35,23,1_10_11,7_55,5,51_67,3,79_83,98,84,2,2_95_46,32_67,8,36_09,4,1,48_65,10_75,2,60_87,71,6,3_46,8,58_54,3,2_95_46,8_24,14_00,18_68,2,19,1_60,2,3_11,8,54_96,2,2_09_20,17,25,1_50_97,3,24,24,0]] , dtype=tf.intaa ) # noqa: E231
# fmt: on
# In 1991 , the remains of Russian Tsar Nicholas II and his family
# ( except for Alexei and Maria ) are discovered .
# The voice of Nicholas's young son , Tsarevich Alexei Nikolaevich , narrates the
# remainder of the story . 1883 Western Siberia ,
# a young Grigori Rasputin is asked by his father and a group of men to perform magic .
# Rasputin has a vision and denounces one of the men as a horse thief . Although his
# father initially slaps him for making such an accusation , Rasputin watches as the
# man is chased outside and beaten . Twenty years later , Rasputin sees a vision of
# the Virgin Mary , prompting him to become a priest . Rasputin quickly becomes famous ,
# with people , even a bishop , begging for his blessing . <eod> </s> <eos>
# fmt: off
snake_case: Any = [33,12_97,2,1,10_09,4,11_09,1_17_39,47_62,3_58,5,25,2_45,22,17_06,17,2_00_98,5,32_15,21,37,11_10,3,13,10_41,4,24,6_03,4_90,2,7_14_77,2_00_98,10_44_47,2,2_09_61,1,26_04,4,1,3_29,3,62_24,8_31,1_60_02,2,8,6_03,7_89_67,2_95_46,23,8_03,20,25,4_16,5,8,2_32,4,2_77,6,18_55,46_01,3,2_95_46,54,8,36_09,5,5_72_11,49,4,1,2_77,18,8,17_55,1_56_91,3,3_41,25,4_16,6_93,4_25_73,71,17,4_01,94,31,1_79_19,2,2_95_46,78_73,18,1,4_35,23,1_10_11,7_55,5,51_67,3,79_83,98,84,2,2_95_46,32_67,8,36_09,4,1,48_65,10_75,2,60_87,71,6,3_46,8,58_54,3,2_95_46,8_24,14_00,18_68,2,19,1_60,2,3_11,8,54_96,2,2_09_20,17,25,1_50_97,3,24,24,0,33,1,18_57,2,1,10_09,4,11_09,1_17_39,47_62,3_58,5,25,2_45,28,11_10,3,13,10_41,4,24,6_03,4_90,2,7_14_77,2_00_98,10_44_47,2,2_09_61,1,26_04,4,1,3_29,3,0] # noqa: E231
# fmt: on
# In 1991, the remains of Russian Tsar Nicholas II and his family (
# except for Alexei and Maria ) are discovered. The voice of young son,
# Tsarevich Alexei Nikolaevich, narrates the remainder of the story.
# 1883 Western Siberia, a young Grigori Rasputin is asked by his father
# and a group of men to perform magic. Rasputin has a vision and
# denounces one of the men as a horse thief. Although his father initially
# slaps him for making such an accusation, Rasputin watches as the man
# is chased outside and beaten. Twenty years later, Rasputin sees a vision
# of the Virgin Mary, prompting him to become a priest.
# Rasputin quickly becomes famous, with people, even a bishop, begging for
# his blessing. <unk> <unk> <eos> In the 1990s, the remains of Russian Tsar
# Nicholas II and his family were discovered. The voice of <unk> young son,
# Tsarevich Alexei Nikolaevich, narrates the remainder of the story.<eos>
snake_case: Tuple = model.generate(SCREAMING_SNAKE_CASE__ , max_length=2_00 , do_sample=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(output_ids[0].numpy().tolist() , SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
import argparse
import json
import os
import fairseq
import torch
from torch import nn
from transformers import (
SpeechaTextaConfig,
SpeechaTextaForCausalLM,
SpeechaTextaTokenizer,
SpeechEncoderDecoderConfig,
SpeechEncoderDecoderModel,
WavaVecaConfig,
WavaVecaFeatureExtractor,
WavaVecaModel,
logging,
)
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"post_extract_proj": "feature_projection.projection",
"encoder.pos_conv.0": "encoder.pos_conv_embed.conv",
"self_attn.k_proj": "encoder.layers.*.attention.k_proj",
"self_attn.v_proj": "encoder.layers.*.attention.v_proj",
"self_attn.q_proj": "encoder.layers.*.attention.q_proj",
"self_attn.out_proj": "encoder.layers.*.attention.out_proj",
"self_attn_layer_norm": "encoder.layers.*.layer_norm",
"fc1": "encoder.layers.*.feed_forward.intermediate_dense",
"fc2": "encoder.layers.*.feed_forward.output_dense",
"final_layer_norm": "encoder.layers.*.final_layer_norm",
"encoder.layer_norm": "encoder.layer_norm",
"w2v_model.layer_norm": "feature_projection.layer_norm",
"quantizer.weight_proj": "quantizer.weight_proj",
"quantizer.vars": "quantizer.codevectors",
"project_q": "project_q",
"final_proj": "project_hid",
"w2v_encoder.proj": "lm_head",
"mask_emb": "masked_spec_embed",
}
__UpperCAmelCase = [
"lm_head",
"quantizer.weight_proj",
"quantizer.codevectors",
"project_q",
"project_hid",
]
def lowerCAmelCase_ ( __A : Any , __A : Optional[Any] , __A : Union[str, Any] , __A : int , __A : Optional[int] ):
'''simple docstring'''
for attribute in key.split('.' ):
snake_case: List[str] = getattr(__A , __A )
if weight_type is not None:
snake_case: Optional[int] = getattr(__A , __A ).shape
else:
snake_case: Optional[int] = hf_pointer.shape
assert hf_shape == value.shape, (
f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
snake_case: Optional[int] = value
elif weight_type == "weight_g":
snake_case: List[str] = value
elif weight_type == "weight_v":
snake_case: Dict = value
elif weight_type == "bias":
snake_case: Optional[Any] = value
else:
snake_case: int = value
logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def lowerCAmelCase_ ( __A : List[Any] , __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = []
snake_case: List[Any] = fairseq_model.state_dict()
snake_case: Union[str, Any] = hf_model.feature_extractor
# if encoder has different dim to decoder -> use proj_weight
snake_case: Dict = None
for name, value in fairseq_dict.items():
snake_case: Tuple = False
if "conv_layers" in name:
load_conv_layer(
__A , __A , __A , __A , hf_model.config.feat_extract_norm == 'group' , )
snake_case: List[Any] = True
elif name.split('.' )[0] == "proj":
snake_case: List[Any] = fairseq_model.proj
snake_case: int = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
snake_case: int = True
if "*" in mapped_key:
snake_case: List[str] = name.split(__A )[0].split('.' )[-2]
snake_case: Dict = mapped_key.replace('*' , __A )
if "weight_g" in name:
snake_case: Tuple = 'weight_g'
elif "weight_v" in name:
snake_case: int = 'weight_v'
elif "bias" in name:
snake_case: Tuple = 'bias'
elif "weight" in name:
snake_case: List[Any] = 'weight'
else:
snake_case: Any = None
set_recursively(__A , __A , __A , __A , __A )
continue
if not is_used:
unused_weights.append(__A )
logger.warning(f"""Unused weights: {unused_weights}""" )
return proj_weight
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : int , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = full_name.split('conv_layers.' )[-1]
snake_case: Tuple = name.split('.' )
snake_case: Any = int(items[0] )
snake_case: Optional[int] = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
snake_case: Tuple = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
snake_case: int = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
snake_case: Any = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
snake_case: str = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(__A )
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case , snake_case: List[Any] = emb.weight.shape
snake_case: Optional[int] = nn.Linear(__A , __A , bias=__A )
snake_case: Any = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
with open(__A , 'r' , encoding='utf-8' ) as f:
snake_case: List[Any] = f.readlines()
snake_case: Any = [line.split(' ' )[0] for line in lines]
snake_case: int = len(__A )
snake_case: Dict = {
'<s>': 0,
'<pad>': 1,
'</s>': 2,
'<unk>': 3,
}
vocab_dict.update(dict(zip(__A , range(4 , num_words + 4 ) ) ) )
return vocab_dict
@torch.no_grad()
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Dict , __A : Any , __A : List[Any] , __A : int , __A : str , ):
'''simple docstring'''
snake_case: Union[str, Any] = WavaVecaConfig.from_pretrained(__A )
snake_case: str = SpeechaTextaConfig.from_pretrained(
__A , vocab_size=__A , decoder_layers=__A , do_stable_layer_norm=__A )
snake_case: List[str] = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=1_60_00 , padding_value=0 , do_normalize=__A , return_attention_mask=__A , )
snake_case , snake_case , snake_case: List[str] = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
snake_case: List[Any] = model[0].eval()
# set weights for wav2vec2 encoder
snake_case: Optional[Any] = WavaVecaModel(__A )
snake_case: Any = recursively_load_weights_wavaveca(model.encoder , __A )
snake_case: Union[str, Any] = SpeechaTextaForCausalLM(__A )
snake_case , snake_case: Optional[Any] = hf_decoder.model.decoder.load_state_dict(model.decoder.state_dict() , strict=__A )
# set output linear layer
unexpected_keys.remove('embed_out' )
snake_case: str = nn.Parameter(model.decoder.embed_out.detach() )
# layer norm is init to identity matrix so leaving it is fine
logger.warning(f"""The following keys are missing when loading the decoder weights: {missing_keys}""" )
logger.warning(f"""The following keys are unexpected when loading the decoder weights: {unexpected_keys}""" )
snake_case: int = SpeechEncoderDecoderModel(encoder=__A , decoder=__A )
snake_case: List[Any] = False
# add projection layer
snake_case: Union[str, Any] = nn.Parameter(projection_layer.weight )
snake_case: Union[str, Any] = nn.Parameter(projection_layer.bias )
snake_case: List[Any] = create_vocab_dict(__A )
with open(os.path.join(__A , 'vocab.json' ) , 'w' ) as fp:
json.dump(__A , __A )
snake_case: Union[str, Any] = SpeechaTextaTokenizer(os.path.join(__A , 'vocab.json' ) )
tokenizer.save_pretrained(__A )
snake_case: Tuple = hf_wavavec.config.to_dict()
snake_case: int = tokenizer.pad_token_id
snake_case: Dict = tokenizer.bos_token_id
snake_case: Optional[int] = tokenizer.eos_token_id
snake_case: Dict = 'speech_to_text_2'
snake_case: Optional[Any] = 'wav2vec2'
snake_case: Tuple = SpeechEncoderDecoderConfig.from_dict(__A )
hf_wavavec.save_pretrained(__A )
feature_extractor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
parser.add_argument("--dict_path", default=None, type=str, help="Path to dict of fine-tuned model")
parser.add_argument(
"--encoder_config_path",
default="facebook/wav2vec2-large-lv60",
type=str,
help="Path to hf encoder wav2vec2 checkpoint config",
)
parser.add_argument(
"--decoder_config_path",
default="facebook/s2t-small-mustc-en-fr-st",
type=str,
help="Path to hf decoder s2t checkpoint config",
)
parser.add_argument("--vocab_size", default=10_224, type=int, help="Vocab size of decoder")
parser.add_argument("--num_decoder_layers", default=7, type=int, help="Number of decoder layers")
__UpperCAmelCase = parser.parse_args()
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.dict_path,
encoder_config_path=args.encoder_config_path,
decoder_config_path=args.decoder_config_path,
vocab_size=args.vocab_size,
num_decoder_layers=args.num_decoder_layers,
) | 692 | 1 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 1_00 ):
'''simple docstring'''
snake_case: List[str] = n * (n + 1) * (2 * n + 1) / 6
snake_case: List[Any] = (n * (n + 1) / 2) ** 2
return int(square_of_sum - sum_of_squares )
if __name__ == "__main__":
print(F'{solution() = }') | 692 | 1 |
'''simple docstring'''
import unittest
import torch
from diffusers import DDIMScheduler, DDPMScheduler, UNetaDModel
from diffusers.training_utils import set_seed
from diffusers.utils.testing_utils import slow
__UpperCAmelCase = False
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__=32 ):
'''simple docstring'''
set_seed(0 )
snake_case: Tuple = UNetaDModel(sample_size=SCREAMING_SNAKE_CASE__ , in_channels=3 , out_channels=3 )
snake_case: Tuple = torch.optim.SGD(model.parameters() , lr=0.00_01 )
return model, optimizer
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = 'cpu' # ensure full determinism without setting the CUBLAS_WORKSPACE_CONFIG env variable
snake_case: Tuple = DDPMScheduler(
num_train_timesteps=10_00 , beta_start=0.00_01 , beta_end=0.02 , beta_schedule='linear' , clip_sample=SCREAMING_SNAKE_CASE__ , )
snake_case: str = DDIMScheduler(
num_train_timesteps=10_00 , beta_start=0.00_01 , beta_end=0.02 , beta_schedule='linear' , clip_sample=SCREAMING_SNAKE_CASE__ , )
assert ddpm_scheduler.config.num_train_timesteps == ddim_scheduler.config.num_train_timesteps
# shared batches for DDPM and DDIM
set_seed(0 )
snake_case: str = [torch.randn((4, 3, 32, 32) ).clip(-1 , 1 ).to(SCREAMING_SNAKE_CASE__ ) for _ in range(4 )]
snake_case: Union[str, Any] = [torch.randn((4, 3, 32, 32) ).to(SCREAMING_SNAKE_CASE__ ) for _ in range(4 )]
snake_case: Dict = [torch.randint(0 , 10_00 , (4,) ).long().to(SCREAMING_SNAKE_CASE__ ) for _ in range(4 )]
# train with a DDPM scheduler
snake_case , snake_case: Union[str, Any] = self.get_model_optimizer(resolution=32 )
model.train().to(SCREAMING_SNAKE_CASE__ )
for i in range(4 ):
optimizer.zero_grad()
snake_case: Any = ddpm_scheduler.add_noise(clean_images[i] , noise[i] , timesteps[i] )
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , timesteps[i] ).sample
snake_case: List[str] = torch.nn.functional.mse_loss(SCREAMING_SNAKE_CASE__ , noise[i] )
loss.backward()
optimizer.step()
del model, optimizer
# recreate the model and optimizer, and retry with DDIM
snake_case , snake_case: Dict = self.get_model_optimizer(resolution=32 )
model.train().to(SCREAMING_SNAKE_CASE__ )
for i in range(4 ):
optimizer.zero_grad()
snake_case: Optional[int] = ddim_scheduler.add_noise(clean_images[i] , noise[i] , timesteps[i] )
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ , timesteps[i] ).sample
snake_case: Tuple = torch.nn.functional.mse_loss(SCREAMING_SNAKE_CASE__ , noise[i] )
loss.backward()
optimizer.step()
del model, optimizer
self.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-5 ) )
self.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-5 ) ) | 692 |
'''simple docstring'''
import argparse
from typing import Dict
import tensorflow as tf
import torch
from tqdm import tqdm
from transformers import BigBirdPegasusConfig, BigBirdPegasusForConditionalGeneration
__UpperCAmelCase = [
# tf -> hf
("/", "."),
("layer_", "layers."),
("kernel", "weight"),
("beta", "bias"),
("gamma", "weight"),
("pegasus", "model"),
]
__UpperCAmelCase = [
(".output.dense", ".fc2"),
("intermediate.LayerNorm", "final_layer_norm"),
("intermediate.dense", "fc1"),
]
__UpperCAmelCase = (
INIT_COMMON
+ [
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.out_proj"),
("attention.self", "self_attn"),
("attention.encdec.LayerNorm", "encoder_attn_layer_norm"),
("attention.encdec_output.dense", "encoder_attn.out_proj"),
("attention.encdec", "encoder_attn"),
("key", "k_proj"),
("value", "v_proj"),
("query", "q_proj"),
("decoder.LayerNorm", "decoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = (
INIT_COMMON
+ [
("embeddings.word_embeddings", "shared.weight"),
("embeddings.position_embeddings", "embed_positions.weight"),
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.output"),
("attention.self", "self_attn.self"),
("encoder.LayerNorm", "encoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = [
"encdec/key/bias",
"encdec/query/bias",
"encdec/value/bias",
"self/key/bias",
"self/query/bias",
"self/value/bias",
"encdec_output/dense/bias",
"attention/output/dense/bias",
]
def lowerCAmelCase_ ( __A : Dict , __A : List[Any] ):
'''simple docstring'''
for tf_name, hf_name in patterns:
snake_case: List[Any] = k.replace(__A , __A )
return k
def lowerCAmelCase_ ( __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[int] = BigBirdPegasusConfig(**__A )
snake_case: List[Any] = BigBirdPegasusForConditionalGeneration(__A )
snake_case: Any = torch_model.state_dict()
snake_case: Any = {}
# separating decoder weights
snake_case: Optional[Any] = {k: tf_weights[k] for k in tf_weights if k.startswith('pegasus/decoder' )}
snake_case: Any = {k: tf_weights[k] for k in tf_weights if not k.startswith('pegasus/decoder' )}
for k, v in tqdm(decoder_weights.items() , 'tf -> hf conversion' ):
snake_case: List[str] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Any = DECODER_PATTERNS
snake_case: int = rename_state_dict_key(__A , __A )
if new_k not in state_dict:
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: Optional[Any] = v.T
snake_case: Any = torch.from_numpy(__A )
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
for k, v in tqdm(remaining_weights.items() , 'tf -> hf conversion' ):
snake_case: List[Any] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Union[str, Any] = REMAINING_PATTERNS
snake_case: str = rename_state_dict_key(__A , __A )
if new_k not in state_dict and k != "pegasus/embeddings/position_embeddings":
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: int = v.T
snake_case: Any = torch.from_numpy(__A )
if k != "pegasus/embeddings/position_embeddings":
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
snake_case: str = mapping['model.embed_positions.weight']
snake_case: Any = mapping.pop('model.embed_positions.weight' )
snake_case , snake_case: Union[str, Any] = torch_model.load_state_dict(__A , strict=__A )
snake_case: Optional[int] = [
k
for k in missing
if k
not in [
'final_logits_bias',
'model.encoder.embed_tokens.weight',
'model.decoder.embed_tokens.weight',
'lm_head.weight',
]
]
assert unexpected_missing == [], f"""no matches found for the following torch keys {unexpected_missing}"""
assert extra == [], f"""no matches found for the following tf keys {extra}"""
return torch_model
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
snake_case: Tuple = tf.train.list_variables(__A )
snake_case: str = {}
snake_case: List[str] = ['global_step']
for name, shape in tqdm(__A , desc='converting tf checkpoint to dict' ):
snake_case: str = any(pat in name for pat in ignore_name )
if skip_key:
continue
snake_case: Any = tf.train.load_variable(__A , __A )
snake_case: Optional[int] = array
return tf_weights
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict ):
'''simple docstring'''
snake_case: int = get_tf_weights_as_numpy(__A )
snake_case: int = convert_bigbird_pegasus(__A , __A )
torch_model.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--tf_ckpt_path", type=str, help="passed to tf.train.list_variables")
parser.add_argument("--save_dir", default=None, type=str, help="Path to the output PyTorch model.")
__UpperCAmelCase = parser.parse_args()
__UpperCAmelCase = {}
convert_bigbird_pegasus_ckpt_to_pytorch(args.tf_ckpt_path, args.save_dir, config_update=config_update) | 692 | 1 |
'''simple docstring'''
import inspect
import re
from transformers.utils import direct_transformers_import
# All paths are set with the intent you should run this script from the root of the repo with the command
# python utils/check_config_docstrings.py
__UpperCAmelCase = "src/transformers"
# This is to make sure the transformers module imported is the one in the repo.
__UpperCAmelCase = direct_transformers_import(PATH_TO_TRANSFORMERS)
__UpperCAmelCase = transformers.models.auto.configuration_auto.CONFIG_MAPPING
# Regex pattern used to find the checkpoint mentioned in the docstring of `config_class`.
# For example, `[bert-base-uncased](https://huggingface.co/bert-base-uncased)`
__UpperCAmelCase = re.compile(R"\[(.+?)\]\((https://huggingface\.co/.+?)\)")
__UpperCAmelCase = {
"DecisionTransformerConfig",
"EncoderDecoderConfig",
"MusicgenConfig",
"RagConfig",
"SpeechEncoderDecoderConfig",
"TimmBackboneConfig",
"VisionEncoderDecoderConfig",
"VisionTextDualEncoderConfig",
"LlamaConfig",
}
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
snake_case: Union[str, Any] = None
# source code of `config_class`
snake_case: Optional[int] = inspect.getsource(__A )
snake_case: str = _re_checkpoint.findall(__A )
# Each `checkpoint` is a tuple of a checkpoint name and a checkpoint link.
# For example, `('bert-base-uncased', 'https://huggingface.co/bert-base-uncased')`
for ckpt_name, ckpt_link in checkpoints:
# allow the link to end with `/`
if ckpt_link.endswith('/' ):
snake_case: Optional[int] = ckpt_link[:-1]
# verify the checkpoint name corresponds to the checkpoint link
snake_case: int = f"""https://huggingface.co/{ckpt_name}"""
if ckpt_link == ckpt_link_from_name:
snake_case: List[str] = ckpt_name
break
return checkpoint
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: int = []
for config_class in list(CONFIG_MAPPING.values() ):
# Skip deprecated models
if "models.deprecated" in config_class.__module__:
continue
snake_case: str = get_checkpoint_from_config_class(__A )
snake_case: Union[str, Any] = config_class.__name__
if checkpoint is None and name not in CONFIG_CLASSES_TO_IGNORE_FOR_DOCSTRING_CHECKPOINT_CHECK:
configs_without_checkpoint.append(__A )
if len(__A ) > 0:
snake_case: Any = '\n'.join(sorted(__A ) )
raise ValueError(f"""The following configurations don't contain any valid checkpoint:\n{message}""" )
if __name__ == "__main__":
check_config_docstrings_have_checkpoints() | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: str = [0] * len(__A )
snake_case: Tuple = []
snake_case: Tuple = [1] * len(__A )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(__A ) ):
if indegree[i] == 0:
queue.append(__A )
while queue:
snake_case: int = queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
snake_case: Any = long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(__A )
print(max(__A ) )
# Adjacency list of Graph
__UpperCAmelCase = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph) | 692 | 1 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
import numpy as np
import pytest
from transformers import BertTokenizer, BertTokenizerFast
from transformers.models.bert.tokenization_bert import VOCAB_FILES_NAMES
from transformers.testing_utils import require_vision
from transformers.utils import FEATURE_EXTRACTOR_NAME, is_vision_available
if is_vision_available():
from PIL import Image
from transformers import ChineseCLIPImageProcessor, ChineseCLIPProcessor
@require_vision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = tempfile.mkdtemp()
snake_case: Optional[Any] = [
'[UNK]',
'[CLS]',
'[SEP]',
'[PAD]',
'[MASK]',
'的',
'价',
'格',
'是',
'15',
'便',
'alex',
'##andra',
'，',
'。',
'-',
't',
'shirt',
]
snake_case: Optional[Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
snake_case: Optional[int] = {
'do_resize': True,
'size': {'height': 2_24, 'width': 2_24},
'do_center_crop': True,
'crop_size': {'height': 18, 'width': 18},
'do_normalize': True,
'image_mean': [0.48_14_54_66, 0.4_57_82_75, 0.40_82_10_73],
'image_std': [0.26_86_29_54, 0.26_13_02_58, 0.27_57_77_11],
'do_convert_rgb': True,
}
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , SCREAMING_SNAKE_CASE__ )
with open(self.image_processor_file , 'w' , encoding='utf-8' ) as fp:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizer.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizerFast.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return ChineseCLIPImageProcessor.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Tuple = [Image.fromarray(np.moveaxis(SCREAMING_SNAKE_CASE__ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizer()
snake_case: Union[str, Any] = self.get_rust_tokenizer()
snake_case: Union[str, Any] = self.get_image_processor()
snake_case: List[str] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_slow.save_pretrained(self.tmpdirname )
snake_case: List[str] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname , use_fast=SCREAMING_SNAKE_CASE__ )
snake_case: Any = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_fast.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor_slow.tokenizer.get_vocab() , tokenizer_slow.get_vocab() )
self.assertEqual(processor_fast.tokenizer.get_vocab() , tokenizer_fast.get_vocab() )
self.assertEqual(tokenizer_slow.get_vocab() , tokenizer_fast.get_vocab() )
self.assertIsInstance(processor_slow.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor_slow.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertEqual(processor_fast.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertIsInstance(processor_slow.image_processor , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=self.get_tokenizer() , image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = self.get_tokenizer(cls_token='(CLS)' , sep_token='(SEP)' )
snake_case: Union[str, Any] = self.get_image_processor(do_normalize=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = ChineseCLIPProcessor.from_pretrained(
self.tmpdirname , cls_token='(CLS)' , sep_token='(SEP)' , do_normalize=SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
self.assertIsInstance(processor.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_image_processor()
snake_case: Tuple = self.get_tokenizer()
snake_case: Optional[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.prepare_image_inputs()
snake_case: List[Any] = image_processor(SCREAMING_SNAKE_CASE__ , return_tensors='np' )
snake_case: Dict = processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='np' )
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_image_processor()
snake_case: Optional[int] = self.get_tokenizer()
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Union[str, Any] = processor(text=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer(SCREAMING_SNAKE_CASE__ )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Tuple = self.prepare_image_inputs()
snake_case: Any = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , ['input_ids', 'token_type_ids', 'attention_mask', 'pixel_values'] )
# test if it raises when no input is passed
with pytest.raises(SCREAMING_SNAKE_CASE__ ):
processor()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.get_image_processor()
snake_case: str = self.get_tokenizer()
snake_case: Union[str, Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = [[1, 4, 5, 8, 1, 0, 8], [3, 4, 3, 1, 1, 8, 9]]
snake_case: int = processor.batch_decode(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = 'Alexandra，T-shirt的价格是15便士。'
snake_case: List[Any] = self.prepare_image_inputs()
snake_case: Dict = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , processor.model_input_names ) | 692 | 1 |
'''simple docstring'''
import unittest
from parameterized import parameterized
from transformers import AutoTokenizer, GPTNeoXConfig, is_torch_available, set_seed
from transformers.testing_utils import require_torch, slow, torch_device
from ...generation.test_utils import GenerationTesterMixin
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import (
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
GPTNeoXModel,
)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=13 , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=99 , SCREAMING_SNAKE_CASE__=64 , SCREAMING_SNAKE_CASE__=5 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=37 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=None , ):
'''simple docstring'''
snake_case: Tuple = parent
snake_case: Optional[Any] = batch_size
snake_case: Any = seq_length
snake_case: Optional[Any] = is_training
snake_case: Union[str, Any] = use_input_mask
snake_case: Tuple = use_token_type_ids
snake_case: List[str] = use_labels
snake_case: Optional[int] = vocab_size
snake_case: Union[str, Any] = hidden_size
snake_case: int = num_hidden_layers
snake_case: Optional[int] = num_attention_heads
snake_case: Tuple = intermediate_size
snake_case: Union[str, Any] = hidden_act
snake_case: str = hidden_dropout_prob
snake_case: Dict = attention_probs_dropout_prob
snake_case: str = max_position_embeddings
snake_case: Optional[int] = type_vocab_size
snake_case: List[str] = type_sequence_label_size
snake_case: str = initializer_range
snake_case: Union[str, Any] = num_labels
snake_case: List[str] = num_choices
snake_case: Any = scope
snake_case: Union[str, Any] = vocab_size - 1
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: Dict = None
if self.use_input_mask:
snake_case: Union[str, Any] = random_attention_mask([self.batch_size, self.seq_length] )
snake_case: List[str] = None
if self.use_labels:
snake_case: str = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
snake_case: Dict = self.get_config()
return config, input_ids, input_mask, token_labels
def _UpperCamelCase ( self ):
'''simple docstring'''
return GPTNeoXConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=SCREAMING_SNAKE_CASE__ , initializer_range=self.initializer_range , pad_token_id=self.pad_token_id , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case , snake_case , snake_case: Optional[Any] = self.prepare_config_and_inputs()
snake_case: Tuple = True
return config, input_ids, input_mask, token_labels
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = GPTNeoXModel(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Dict = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )
snake_case: int = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = True
snake_case: str = GPTNeoXModel(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = GPTNeoXForCausalLM(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.num_labels
snake_case: Tuple = GPTNeoXForQuestionAnswering(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.num_labels
snake_case: List[Any] = GPTNeoXForSequenceClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Dict = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case: str = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Tuple = self.num_labels
snake_case: List[Any] = GPTNeoXForTokenClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: int = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Union[str, Any] = True
snake_case: Union[str, Any] = GPTNeoXForCausalLM(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
# first forward pass
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , use_cache=SCREAMING_SNAKE_CASE__ )
snake_case: str = outputs.past_key_values
# create hypothetical multiple next token and extent to next_input_ids
snake_case: Tuple = ids_tensor((self.batch_size, 3) , config.vocab_size )
snake_case: Tuple = ids_tensor((self.batch_size, 3) , vocab_size=2 )
# append to next input_ids and
snake_case: List[Any] = torch.cat([input_ids, next_tokens] , dim=-1 )
snake_case: Optional[int] = torch.cat([input_mask, next_mask] , dim=-1 )
snake_case: List[str] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ )
snake_case: Any = output_from_no_past['hidden_states'][0]
snake_case: Optional[Any] = model(
SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , past_key_values=SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , )['hidden_states'][0]
# select random slice
snake_case: Union[str, Any] = ids_tensor((1,) , output_from_past.shape[-1] ).item()
snake_case: Tuple = output_from_no_past[:, -3:, random_slice_idx].detach()
snake_case: List[Any] = output_from_past[:, :, random_slice_idx].detach()
self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1] )
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-3 ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.prepare_config_and_inputs()
snake_case , snake_case , snake_case , snake_case: Any = config_and_inputs
snake_case: List[str] = {'input_ids': input_ids, 'attention_mask': input_mask}
return config, inputs_dict
@require_torch
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (
(
GPTNeoXModel,
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
)
if is_torch_available()
else ()
)
__UpperCamelCase = (GPTNeoXForCausalLM,) if is_torch_available() else ()
__UpperCamelCase = (
{
"feature-extraction": GPTNeoXModel,
"question-answering": GPTNeoXForQuestionAnswering,
"text-classification": GPTNeoXForSequenceClassification,
"text-generation": GPTNeoXForCausalLM,
"token-classification": GPTNeoXForTokenClassification,
"zero-shot": GPTNeoXForSequenceClassification,
}
if is_torch_available()
else {}
)
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = GPTNeoXModelTester(self )
snake_case: Union[str, Any] = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , hidden_size=64 , num_attention_heads=8 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case , snake_case , snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case , snake_case , snake_case: Any = self.model_tester.prepare_config_and_inputs_for_decoder()
self.model_tester.create_and_check_model_as_decoder(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case , snake_case , snake_case: List[str] = self.model_tester.prepare_config_and_inputs_for_decoder()
snake_case: List[str] = None
self.model_tester.create_and_check_model_as_decoder(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case , snake_case , snake_case: List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_decoder_model_past_large_inputs(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_causal_lm(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_question_answering(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_sequence_classification(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_token_classification(*SCREAMING_SNAKE_CASE__ )
@unittest.skip(reason='Feed forward chunking is not implemented' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@parameterized.expand([('linear',), ('dynamic',)] )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: Optional[Any] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[str] = ids_tensor([1, 10] , config.vocab_size )
snake_case: Optional[Any] = ids_tensor([1, int(config.max_position_embeddings * 1.5 )] , config.vocab_size )
set_seed(42 ) # Fixed seed at init time so the two models get the same random weights
snake_case: List[str] = GPTNeoXModel(SCREAMING_SNAKE_CASE__ )
original_model.to(SCREAMING_SNAKE_CASE__ )
original_model.eval()
snake_case: Dict = original_model(SCREAMING_SNAKE_CASE__ ).last_hidden_state
snake_case: List[str] = original_model(SCREAMING_SNAKE_CASE__ ).last_hidden_state
set_seed(42 ) # Fixed seed at init time so the two models get the same random weights
snake_case: Any = {'type': scaling_type, 'factor': 10.0}
snake_case: Optional[int] = GPTNeoXModel(SCREAMING_SNAKE_CASE__ )
scaled_model.to(SCREAMING_SNAKE_CASE__ )
scaled_model.eval()
snake_case: str = scaled_model(SCREAMING_SNAKE_CASE__ ).last_hidden_state
snake_case: Dict = scaled_model(SCREAMING_SNAKE_CASE__ ).last_hidden_state
# Dynamic scaling does not change the RoPE embeddings until it receives an input longer than the original
# maximum sequence length, so the outputs for the short input should match.
if scaling_type == "dynamic":
self.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-5 ) )
else:
self.assertFalse(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-5 ) )
# The output should be different for long inputs
self.assertFalse(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-5 ) )
@require_torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = AutoTokenizer.from_pretrained('EleutherAI/pythia-410m-deduped' )
for checkpointing in [True, False]:
snake_case: List[Any] = GPTNeoXForCausalLM.from_pretrained('EleutherAI/pythia-410m-deduped' )
if checkpointing:
model.gradient_checkpointing_enable()
else:
model.gradient_checkpointing_disable()
model.to(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer('My favorite food is' , return_tensors='pt' ).to(SCREAMING_SNAKE_CASE__ )
# The hub repo. is updated on 2023-04-04, resulting in poor outputs.
# See: https://github.com/huggingface/transformers/pull/24193
snake_case: Dict = 'My favorite food is a good old-fashioned, old-fashioned, old-fashioned.\n\nI\'m not sure'
snake_case: Dict = model.generate(**SCREAMING_SNAKE_CASE__ , do_sample=SCREAMING_SNAKE_CASE__ , max_new_tokens=20 )
snake_case: Union[str, Any] = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ )[0]
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/swinv2-tiny-patch4-window8-256": (
"https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256/resolve/main/config.json"
),
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "swinv2"
__UpperCamelCase = {
"num_attention_heads": "num_heads",
"num_hidden_layers": "num_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=96 , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[3, 6, 12, 24] , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=32 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: int = image_size
snake_case: Union[str, Any] = patch_size
snake_case: List[str] = num_channels
snake_case: Tuple = embed_dim
snake_case: str = depths
snake_case: Any = len(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = num_heads
snake_case: Optional[int] = window_size
snake_case: Any = mlp_ratio
snake_case: Optional[int] = qkv_bias
snake_case: Union[str, Any] = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: Dict = drop_path_rate
snake_case: List[str] = hidden_act
snake_case: int = use_absolute_embeddings
snake_case: Any = layer_norm_eps
snake_case: Dict = initializer_range
snake_case: List[Any] = encoder_stride
# we set the hidden_size attribute in order to make Swinv2 work with VisionEncoderDecoderModel
# this indicates the channel dimension after the last stage of the model
snake_case: Tuple = int(embed_dim * 2 ** (len(SCREAMING_SNAKE_CASE__ ) - 1) )
snake_case: Union[str, Any] = (0, 0, 0, 0) | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = "\n# Transformers installation\n! pip install transformers datasets\n# To install from source instead of the last release, comment the command above and uncomment the following one.\n# ! pip install git+https://github.com/huggingface/transformers.git\n"
__UpperCAmelCase = [{"type": "code", "content": INSTALL_CONTENT}]
__UpperCAmelCase = {
"{processor_class}": "FakeProcessorClass",
"{model_class}": "FakeModelClass",
"{object_class}": "FakeObjectClass",
} | 692 |
'''simple docstring'''
import os
import sys
import unittest
__UpperCAmelCase = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__))))
sys.path.append(os.path.join(git_repo_path, "utils"))
import check_dummies # noqa: E402
from check_dummies import create_dummy_files, create_dummy_object, find_backend, read_init # noqa: E402
# Align TRANSFORMERS_PATH in check_dummies with the current path
__UpperCAmelCase = os.path.join(git_repo_path, "src", "transformers")
__UpperCAmelCase = "\n{0} = None\n"
__UpperCAmelCase = "\nclass {0}(metaclass=DummyObject):\n _backends = {1}\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, {1})\n"
__UpperCAmelCase = "\ndef {0}(*args, **kwargs):\n requires_backends({0}, {1})\n"
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = find_backend(' _import_structure["models.albert"].append("AlbertTokenizerFast")' )
self.assertIsNone(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = find_backend(' if not is_tokenizers_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tokenizers' )
snake_case: List[Any] = find_backend(' if not is_tensorflow_text_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tensorflow_text' )
snake_case: int = find_backend(' if not (is_sentencepiece_available() and is_tokenizers_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers' )
snake_case: Optional[Any] = find_backend(
' if not (is_sentencepiece_available() and is_tensorflow_text_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tensorflow_text' )
snake_case: Dict = find_backend(
' if not (is_sentencepiece_available() and is_tokenizers_available() and is_vision_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers_and_vision' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = read_init()
# We don't assert on the exact list of keys to allow for smooth grow of backend-specific objects
self.assertIn('torch' , SCREAMING_SNAKE_CASE__ )
self.assertIn('tensorflow_text' , SCREAMING_SNAKE_CASE__ )
self.assertIn('sentencepiece_and_tokenizers' , SCREAMING_SNAKE_CASE__ )
# Likewise, we can't assert on the exact content of a key
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertModel' , objects['tf'] )
self.assertIn('FlaxBertModel' , objects['flax'] )
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertTokenizer' , objects['tensorflow_text'] )
self.assertIn('convert_slow_tokenizer' , objects['sentencepiece_and_tokenizers'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = create_dummy_object('CONSTANT' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , '\nCONSTANT = None\n' )
snake_case: Any = create_dummy_object('function' , '\'torch\'' )
self.assertEqual(
SCREAMING_SNAKE_CASE__ , '\ndef function(*args, **kwargs):\n requires_backends(function, \'torch\')\n' )
snake_case: Optional[int] = '\nclass FakeClass(metaclass=DummyObject):\n _backends = \'torch\'\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, \'torch\')\n'
snake_case: Tuple = create_dummy_object('FakeClass' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = '# This file is autogenerated by the command `make fix-copies`, do not edit.\nfrom ..utils import DummyObject, requires_backends\n\n\nCONSTANT = None\n\n\ndef function(*args, **kwargs):\n requires_backends(function, ["torch"])\n\n\nclass FakeClass(metaclass=DummyObject):\n _backends = ["torch"]\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, ["torch"])\n'
snake_case: Optional[int] = create_dummy_files({'torch': ['CONSTANT', 'function', 'FakeClass']} )
self.assertEqual(dummy_files['torch'] , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
import string
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
for key in range(len(string.ascii_uppercase ) ):
snake_case: Tuple = ''
for symbol in message:
if symbol in string.ascii_uppercase:
snake_case: int = string.ascii_uppercase.find(__A )
snake_case: List[Any] = num - key
if num < 0:
snake_case: Tuple = num + len(string.ascii_uppercase )
snake_case: str = translated + string.ascii_uppercase[num]
else:
snake_case: Optional[int] = translated + symbol
print(f"""Decryption using Key #{key}: {translated}""" )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: str = input('Encrypted message: ' )
snake_case: str = message.upper()
decrypt(__A )
if __name__ == "__main__":
import doctest
doctest.testmod()
main() | 692 |
'''simple docstring'''
import os
import warnings
from typing import List, Optional
from ...tokenization_utils_base import BatchEncoding
from ...utils import logging
from .configuration_rag import RagConfig
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = question_encoder
snake_case: Union[str, Any] = generator
snake_case: Optional[int] = self.question_encoder
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if os.path.isfile(SCREAMING_SNAKE_CASE__ ):
raise ValueError(F"""Provided path ({save_directory}) should be a directory, not a file""" )
os.makedirs(SCREAMING_SNAKE_CASE__ , exist_ok=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'question_encoder_tokenizer' )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'generator_tokenizer' )
self.question_encoder.save_pretrained(SCREAMING_SNAKE_CASE__ )
self.generator.save_pretrained(SCREAMING_SNAKE_CASE__ )
@classmethod
def _UpperCamelCase ( cls , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
from ..auto.tokenization_auto import AutoTokenizer
snake_case: int = kwargs.pop('config' , SCREAMING_SNAKE_CASE__ )
if config is None:
snake_case: str = RagConfig.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.question_encoder , subfolder='question_encoder_tokenizer' )
snake_case: Dict = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.generator , subfolder='generator_tokenizer' )
return cls(question_encoder=SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ )
def __call__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.current_tokenizer(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.batch_decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.question_encoder
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.generator
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "longest" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
warnings.warn(
'`prepare_seq2seq_batch` is deprecated and will be removed in version 5 of 🤗 Transformers. Use the '
'regular `__call__` method to prepare your inputs and the tokenizer under the `with_target_tokenizer` '
'context manager to prepare your targets. See the documentation of your specific tokenizer for more '
'details' , SCREAMING_SNAKE_CASE__ , )
if max_length is None:
snake_case: Optional[Any] = self.current_tokenizer.model_max_length
snake_case: int = self(
SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if tgt_texts is None:
return model_inputs
# Process tgt_texts
if max_target_length is None:
snake_case: Any = self.current_tokenizer.model_max_length
snake_case: List[str] = self(
text_target=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: Dict = labels['input_ids']
return model_inputs | 692 | 1 |
'''simple docstring'''
import torch
from diffusers import CMStochasticIterativeScheduler
from .test_schedulers import SchedulerCommonTest
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = (CMStochasticIterativeScheduler,)
__UpperCamelCase = 10
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = {
'num_train_timesteps': 2_01,
'sigma_min': 0.0_02,
'sigma_max': 80.0,
}
config.update(**SCREAMING_SNAKE_CASE__ )
return config
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = 10
snake_case: Optional[Any] = self.get_scheduler_config()
snake_case: Union[str, Any] = self.scheduler_classes[0](**SCREAMING_SNAKE_CASE__ )
scheduler.set_timesteps(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = scheduler.timesteps[0]
snake_case: Dict = scheduler.timesteps[1]
snake_case: Optional[Any] = self.dummy_sample
snake_case: str = 0.1 * sample
snake_case: int = scheduler.step(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ).prev_sample
snake_case: int = scheduler.step(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ).prev_sample
self.assertEqual(output_a.shape , sample.shape )
self.assertEqual(output_a.shape , output_a.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
for timesteps in [10, 50, 1_00, 10_00]:
self.check_over_configs(num_train_timesteps=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
for clip_denoised in [True, False]:
self.check_over_configs(clip_denoised=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.scheduler_classes[0]
snake_case: List[Any] = self.get_scheduler_config()
snake_case: Tuple = scheduler_class(**SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = 1
scheduler.set_timesteps(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = scheduler.timesteps
snake_case: Optional[int] = torch.manual_seed(0 )
snake_case: Any = self.dummy_model()
snake_case: List[str] = self.dummy_sample_deter * scheduler.init_noise_sigma
for i, t in enumerate(SCREAMING_SNAKE_CASE__ ):
# 1. scale model input
snake_case: Optional[int] = scheduler.scale_model_input(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# 2. predict noise residual
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# 3. predict previous sample x_t-1
snake_case: Union[str, Any] = scheduler.step(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ ).prev_sample
snake_case: Union[str, Any] = pred_prev_sample
snake_case: str = torch.sum(torch.abs(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = torch.mean(torch.abs(SCREAMING_SNAKE_CASE__ ) )
assert abs(result_sum.item() - 1_92.76_14 ) < 1E-2
assert abs(result_mean.item() - 0.25_10 ) < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.scheduler_classes[0]
snake_case: Optional[Any] = self.get_scheduler_config()
snake_case: int = scheduler_class(**SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = [1_06, 0]
scheduler.set_timesteps(timesteps=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = scheduler.timesteps
snake_case: str = torch.manual_seed(0 )
snake_case: int = self.dummy_model()
snake_case: List[str] = self.dummy_sample_deter * scheduler.init_noise_sigma
for t in timesteps:
# 1. scale model input
snake_case: List[Any] = scheduler.scale_model_input(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# 2. predict noise residual
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# 3. predict previous sample x_t-1
snake_case: str = scheduler.step(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ ).prev_sample
snake_case: str = pred_prev_sample
snake_case: Dict = torch.sum(torch.abs(SCREAMING_SNAKE_CASE__ ) )
snake_case: str = torch.mean(torch.abs(SCREAMING_SNAKE_CASE__ ) )
assert abs(result_sum.item() - 3_47.63_57 ) < 1E-2
assert abs(result_mean.item() - 0.45_27 ) < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.scheduler_classes[0]
snake_case: Tuple = self.get_scheduler_config()
snake_case: List[Any] = scheduler_class(**SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = [39, 30, 12, 15, 0]
with self.assertRaises(SCREAMING_SNAKE_CASE__ , msg='`timesteps` must be in descending order.' ):
scheduler.set_timesteps(timesteps=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.scheduler_classes[0]
snake_case: int = self.get_scheduler_config()
snake_case: Optional[Any] = scheduler_class(**SCREAMING_SNAKE_CASE__ )
snake_case: int = [39, 30, 12, 1, 0]
snake_case: Union[str, Any] = len(SCREAMING_SNAKE_CASE__ )
with self.assertRaises(SCREAMING_SNAKE_CASE__ , msg='Can only pass one of `num_inference_steps` or `timesteps`.' ):
scheduler.set_timesteps(num_inference_steps=SCREAMING_SNAKE_CASE__ , timesteps=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.scheduler_classes[0]
snake_case: List[str] = self.get_scheduler_config()
snake_case: List[str] = scheduler_class(**SCREAMING_SNAKE_CASE__ )
snake_case: str = [scheduler.config.num_train_timesteps]
with self.assertRaises(
SCREAMING_SNAKE_CASE__ , msg='`timesteps` must start before `self.config.train_timesteps`: {scheduler.config.num_train_timesteps}}' , ):
scheduler.set_timesteps(timesteps=SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
import importlib
import os
import fsspec
import pytest
from fsspec import register_implementation
from fsspec.registry import _registry as _fsspec_registry
from datasets.filesystems import COMPRESSION_FILESYSTEMS, HfFileSystem, extract_path_from_uri, is_remote_filesystem
from .utils import require_lza, require_zstandard
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
assert "mock" in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert "mock" not in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'mock-s3-bucket'
snake_case: int = f"""s3://{mock_bucket}"""
snake_case: Any = extract_path_from_uri(__A )
assert dataset_path.startswith('s3://' ) is False
snake_case: Union[str, Any] = './local/path'
snake_case: Union[str, Any] = extract_path_from_uri(__A )
assert dataset_path == new_dataset_path
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
snake_case: List[str] = is_remote_filesystem(__A )
assert is_remote is True
snake_case: int = fsspec.filesystem('file' )
snake_case: int = is_remote_filesystem(__A )
assert is_remote is False
@pytest.mark.parametrize('compression_fs_class' , __A )
def lowerCAmelCase_ ( __A : Optional[int] , __A : int , __A : str , __A : Optional[Any] , __A : List[str] , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: Optional[Any] = {'gzip': gz_file, 'xz': xz_file, 'zstd': zstd_file, 'bz2': bza_file, 'lz4': lza_file}
snake_case: Optional[int] = input_paths[compression_fs_class.protocol]
if input_path is None:
snake_case: str = f"""for '{compression_fs_class.protocol}' compression protocol, """
if compression_fs_class.protocol == "lz4":
reason += require_lza.kwargs["reason"]
elif compression_fs_class.protocol == "zstd":
reason += require_zstandard.kwargs["reason"]
pytest.skip(__A )
snake_case: List[str] = fsspec.filesystem(compression_fs_class.protocol , fo=__A )
assert isinstance(__A , __A )
snake_case: Any = os.path.basename(__A )
snake_case: int = expected_filename[: expected_filename.rindex('.' )]
assert fs.glob('*' ) == [expected_filename]
with fs.open(__A , 'r' , encoding='utf-8' ) as f, open(__A , encoding='utf-8' ) as expected_file:
assert f.read() == expected_file.read()
@pytest.mark.parametrize('protocol' , ['zip', 'gzip'] )
def lowerCAmelCase_ ( __A : Any , __A : int , __A : int ):
'''simple docstring'''
snake_case: List[str] = {'zip': zip_jsonl_path, 'gzip': jsonl_gz_path}
snake_case: str = compressed_file_paths[protocol]
snake_case: Dict = 'dataset.jsonl'
snake_case: Optional[Any] = f"""{protocol}://{member_file_path}::{compressed_file_path}"""
snake_case , *snake_case: List[Any] = fsspec.get_fs_token_paths(__A )
assert fs.isfile(__A )
assert not fs.isfile('non_existing_' + member_file_path )
@pytest.mark.integration
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Union[str, Any] , __A : List[Any] ):
'''simple docstring'''
snake_case: Tuple = hf_api.dataset_info(__A , token=__A )
snake_case: List[str] = HfFileSystem(repo_info=__A , token=__A )
assert sorted(hffs.glob('*' ) ) == [".gitattributes", "data"]
assert hffs.isdir('data' )
assert hffs.isfile('.gitattributes' ) and hffs.isfile('data/text_data.txt' )
with open(__A ) as f:
assert hffs.open('data/text_data.txt' , 'r' ).read() == f.read()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'bz2'
# Import module
import datasets.filesystems
# Overwrite protocol and reload
register_implementation(__A , __A , clobber=__A )
with pytest.warns(__A ) as warning_info:
importlib.reload(datasets.filesystems )
assert len(__A ) == 1
assert (
str(warning_info[0].message )
== f"""A filesystem protocol was already set for {protocol} and will be overwritten."""
) | 692 | 1 |
'''simple docstring'''
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTConfig, DeiTForImageClassificationWithTeacher, DeiTImageProcessor
from transformers.utils import logging
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
def lowerCAmelCase_ ( __A : Optional[int] , __A : int=False ):
'''simple docstring'''
snake_case: str = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""deit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""deit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""deit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""deit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""deit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""deit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""deit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""deit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""deit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""deit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
('cls_token', 'deit.embeddings.cls_token'),
('dist_token', 'deit.embeddings.distillation_token'),
('patch_embed.proj.weight', 'deit.embeddings.patch_embeddings.projection.weight'),
('patch_embed.proj.bias', 'deit.embeddings.patch_embeddings.projection.bias'),
('pos_embed', 'deit.embeddings.position_embeddings'),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
('norm.weight', 'layernorm.weight'),
('norm.bias', 'layernorm.bias'),
('pre_logits.fc.weight', 'pooler.dense.weight'),
('pre_logits.fc.bias', 'pooler.dense.bias'),
] )
# if just the base model, we should remove "deit" from all keys that start with "deit"
snake_case: str = [(pair[0], pair[1][4:]) if pair[1].startswith('deit' ) else pair for pair in rename_keys]
else:
# layernorm + classification heads
rename_keys.extend(
[
('norm.weight', 'deit.layernorm.weight'),
('norm.bias', 'deit.layernorm.bias'),
('head.weight', 'cls_classifier.weight'),
('head.bias', 'cls_classifier.bias'),
('head_dist.weight', 'distillation_classifier.weight'),
('head_dist.bias', 'distillation_classifier.bias'),
] )
return rename_keys
def lowerCAmelCase_ ( __A : Optional[Any] , __A : List[str] , __A : Tuple=False ):
'''simple docstring'''
for i in range(config.num_hidden_layers ):
if base_model:
snake_case: List[Any] = ''
else:
snake_case: Optional[Any] = 'deit.'
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case: Dict = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case: Union[str, Any] = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case: str = in_proj_weight[
: config.hidden_size, :
]
snake_case: int = in_proj_bias[: config.hidden_size]
snake_case: List[Any] = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case: int = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case: List[Any] = in_proj_weight[
-config.hidden_size :, :
]
snake_case: List[str] = in_proj_bias[-config.hidden_size :]
def lowerCAmelCase_ ( __A : Dict , __A : Union[str, Any] , __A : List[str] ):
'''simple docstring'''
snake_case: Union[str, Any] = dct.pop(__A )
snake_case: Optional[Any] = val
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: int = 'http://images.cocodataset.org/val2017/000000039769.jpg'
snake_case: Union[str, Any] = Image.open(requests.get(__A , stream=__A ).raw )
return im
@torch.no_grad()
def lowerCAmelCase_ ( __A : Union[str, Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: str = DeiTConfig()
# all deit models have fine-tuned heads
snake_case: Union[str, Any] = False
# dataset (fine-tuned on ImageNet 2012), patch_size and image_size
snake_case: Any = 10_00
snake_case: List[str] = 'huggingface/label-files'
snake_case: Dict = 'imagenet-1k-id2label.json'
snake_case: List[str] = json.load(open(hf_hub_download(__A , __A , repo_type='dataset' ) , 'r' ) )
snake_case: List[str] = {int(__A ): v for k, v in idalabel.items()}
snake_case: Tuple = idalabel
snake_case: Union[str, Any] = {v: k for k, v in idalabel.items()}
snake_case: List[Any] = int(deit_name[-6:-4] )
snake_case: Optional[Any] = int(deit_name[-3:] )
# size of the architecture
if deit_name[9:].startswith('tiny' ):
snake_case: List[str] = 1_92
snake_case: str = 7_68
snake_case: Union[str, Any] = 12
snake_case: int = 3
elif deit_name[9:].startswith('small' ):
snake_case: Optional[int] = 3_84
snake_case: List[str] = 15_36
snake_case: int = 12
snake_case: Optional[Any] = 6
if deit_name[9:].startswith('base' ):
pass
elif deit_name[4:].startswith('large' ):
snake_case: List[Any] = 10_24
snake_case: int = 40_96
snake_case: str = 24
snake_case: Dict = 16
# load original model from timm
snake_case: List[str] = timm.create_model(__A , pretrained=__A )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case: Dict = timm_model.state_dict()
snake_case: Union[str, Any] = create_rename_keys(__A , __A )
for src, dest in rename_keys:
rename_key(__A , __A , __A )
read_in_q_k_v(__A , __A , __A )
# load HuggingFace model
snake_case: List[Any] = DeiTForImageClassificationWithTeacher(__A ).eval()
model.load_state_dict(__A )
# Check outputs on an image, prepared by DeiTImageProcessor
snake_case: List[Any] = int(
(2_56 / 2_24) * config.image_size ) # to maintain same ratio w.r.t. 224 images, see https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L103
snake_case: Tuple = DeiTImageProcessor(size=__A , crop_size=config.image_size )
snake_case: Optional[Any] = image_processor(images=prepare_img() , return_tensors='pt' )
snake_case: List[str] = encoding['pixel_values']
snake_case: List[Any] = model(__A )
snake_case: Tuple = timm_model(__A )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(__A , outputs.logits , atol=1E-3 )
Path(__A ).mkdir(exist_ok=__A )
print(f"""Saving model {deit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--deit_name",
default="vit_deit_base_distilled_patch16_224",
type=str,
help="Name of the DeiT timm model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model directory."
)
__UpperCAmelCase = parser.parse_args()
convert_deit_checkpoint(args.deit_name, args.pytorch_dump_folder_path) | 692 |
'''simple docstring'''
import logging
import os
import random
import sys
from dataclasses import dataclass, field
from typing import Optional
import datasets
import numpy as np
import pandas as pd
from datasets import load_dataset
import transformers
from transformers import (
AutoConfig,
BartForSequenceClassification,
DataCollatorWithPadding,
EvalPrediction,
HfArgumentParser,
TapexTokenizer,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.17.0.dev0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
__UpperCAmelCase = logging.getLogger(__name__)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The name of the dataset to use (via the datasets library)."} )
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} , )
__UpperCamelCase = field(
default=1024 , metadata={
"help": (
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Overwrite the cached preprocessed datasets or not."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Whether to pad all samples to `max_seq_length`. "
"If False, will pad the samples dynamically when batching to the maximum length in the batch."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of prediction examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the training data."} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the validation data."} )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A csv or a json file containing the test data."} )
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.dataset_name is not None:
pass
elif self.train_file is None or self.validation_file is None:
raise ValueError('Need either a GLUE task, a training/validation file or a dataset name.' )
else:
snake_case: str = self.train_file.split('.' )[-1]
assert train_extension in ["csv", "json"], "`train_file` should be a csv or a json file."
snake_case: Optional[Any] = self.validation_file.split('.' )[-1]
assert (
validation_extension == train_extension
), "`validation_file` should have the same extension (csv or json) as `train_file`."
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."} , )
__UpperCamelCase = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case , snake_case , snake_case: Union[str, Any] = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case , snake_case , snake_case: str = parser.parse_args_into_dataclasses()
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
snake_case: Tuple = training_args.get_process_log_level()
logger.setLevel(__A )
datasets.utils.logging.set_verbosity(__A )
transformers.utils.logging.set_verbosity(__A )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"""
+ f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" )
logger.info(f"""Training/evaluation parameters {training_args}""" )
# Detecting last checkpoint.
snake_case: Any = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case: List[Any] = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"""Output directory ({training_args.output_dir}) already exists and is not empty. """
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Set seed before initializing model.
set_seed(training_args.seed )
# Get the datasets: you can either provide your own CSV/JSON training and evaluation files (see below)
# or specify a GLUE benchmark task (the dataset will be downloaded automatically from the datasets Hub).
#
# For JSON files, this script will use the `question` column for the input question and `table` column for the corresponding table.
#
# If the CSVs/JSONs contain only one non-label column, the script does single sentence classification on this
# single column. You can easily tweak this behavior (see below)
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
snake_case: int = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from your local files.
# CSV/JSON training and evaluation files are needed.
snake_case: Optional[int] = {'train': data_args.train_file, 'validation': data_args.validation_file}
# Get the test dataset: you can provide your own CSV/JSON test file (see below)
# when you use `do_predict` without specifying a GLUE benchmark task.
if training_args.do_predict:
if data_args.test_file is not None:
snake_case: Tuple = data_args.train_file.split('.' )[-1]
snake_case: Union[str, Any] = data_args.test_file.split('.' )[-1]
assert (
test_extension == train_extension
), "`test_file` should have the same extension (csv or json) as `train_file`."
snake_case: Union[str, Any] = data_args.test_file
else:
raise ValueError('Need either a GLUE task or a test file for `do_predict`.' )
for key in data_files.keys():
logger.info(f"""load a local file for {key}: {data_files[key]}""" )
if data_args.train_file.endswith('.csv' ):
# Loading a dataset from local csv files
snake_case: List[Any] = load_dataset('csv' , data_files=__A , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from local json files
snake_case: Optional[Any] = load_dataset('json' , data_files=__A , cache_dir=model_args.cache_dir )
# See more about loading any type of standard or custom dataset at
# https://huggingface.co/docs/datasets/loading_datasets.html.
# Labels
snake_case: Tuple = raw_datasets['train'].features['label'].names
snake_case: List[str] = len(__A )
# Load pretrained model and tokenizer
#
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case: Tuple = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# load tapex tokenizer
snake_case: List[str] = TapexTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , add_prefix_space=__A , )
snake_case: Union[str, Any] = BartForSequenceClassification.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# Padding strategy
if data_args.pad_to_max_length:
snake_case: int = 'max_length'
else:
# We will pad later, dynamically at batch creation, to the max sequence length in each batch
snake_case: Union[str, Any] = False
# Some models have set the order of the labels to use, so let's make sure we do use it.
snake_case: Optional[Any] = {'Refused': 0, 'Entailed': 1}
snake_case: List[Any] = {0: 'Refused', 1: 'Entailed'}
if data_args.max_seq_length > tokenizer.model_max_length:
logger.warning(
f"""The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"""
f"""model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}.""" )
snake_case: List[str] = min(data_args.max_seq_length , tokenizer.model_max_length )
def preprocess_tabfact_function(__A : Any ):
# Tokenize the texts
def _convert_table_text_to_pandas(__A : Dict ):
snake_case: str = [_table_row.split('#' ) for _table_row in _table_text.strip('\n' ).split('\n' )]
snake_case: List[str] = pd.DataFrame.from_records(_table_content[1:] , columns=_table_content[0] )
return _table_pd
snake_case: str = examples['statement']
snake_case: int = list(map(_convert_table_text_to_pandas , examples['table_text'] ) )
snake_case: List[Any] = tokenizer(__A , __A , padding=__A , max_length=__A , truncation=__A )
snake_case: List[Any] = examples['label']
return result
with training_args.main_process_first(desc='dataset map pre-processing' ):
snake_case: int = raw_datasets.map(
__A , batched=__A , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on dataset' , )
if training_args.do_train:
if "train" not in raw_datasets:
raise ValueError('--do_train requires a train dataset' )
snake_case: List[str] = raw_datasets['train']
if data_args.max_train_samples is not None:
snake_case: Tuple = train_dataset.select(range(data_args.max_train_samples ) )
if training_args.do_eval:
if "validation" not in raw_datasets and "validation_matched" not in raw_datasets:
raise ValueError('--do_eval requires a validation dataset' )
snake_case: Any = raw_datasets['validation']
if data_args.max_eval_samples is not None:
snake_case: Optional[int] = eval_dataset.select(range(data_args.max_eval_samples ) )
if training_args.do_predict or data_args.test_file is not None:
if "test" not in raw_datasets and "test_matched" not in raw_datasets:
raise ValueError('--do_predict requires a test dataset' )
snake_case: str = raw_datasets['test']
if data_args.max_predict_samples is not None:
snake_case: List[str] = predict_dataset.select(range(data_args.max_predict_samples ) )
# Log a few random samples from the training set:
if training_args.do_train:
for index in random.sample(range(len(__A ) ) , 3 ):
logger.info(f"""Sample {index} of the training set: {train_dataset[index]}.""" )
# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def compute_metrics(__A : EvalPrediction ):
snake_case: int = p.predictions[0] if isinstance(p.predictions , __A ) else p.predictions
snake_case: List[str] = np.argmax(__A , axis=1 )
return {"accuracy": (preds == p.label_ids).astype(np.floataa ).mean().item()}
# Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
if data_args.pad_to_max_length:
snake_case: str = default_data_collator
elif training_args.fpaa:
snake_case: List[str] = DataCollatorWithPadding(__A , pad_to_multiple_of=8 )
else:
snake_case: List[Any] = None
# Initialize our Trainer
snake_case: List[str] = Trainer(
model=__A , args=__A , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , compute_metrics=__A , tokenizer=__A , data_collator=__A , )
# Training
if training_args.do_train:
snake_case: Optional[int] = None
if training_args.resume_from_checkpoint is not None:
snake_case: str = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
snake_case: Optional[Any] = last_checkpoint
snake_case: Union[str, Any] = trainer.train(resume_from_checkpoint=__A )
snake_case: List[Any] = train_result.metrics
snake_case: List[Any] = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(__A )
)
snake_case: Optional[Any] = min(__A , len(__A ) )
trainer.save_model() # Saves the tokenizer too for easy upload
trainer.log_metrics('train' , __A )
trainer.save_metrics('train' , __A )
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info('*** Evaluate ***' )
snake_case: Dict = trainer.evaluate(eval_dataset=__A )
snake_case: Optional[int] = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(__A )
snake_case: Dict = min(__A , len(__A ) )
trainer.log_metrics('eval' , __A )
trainer.save_metrics('eval' , __A )
if training_args.do_predict:
logger.info('*** Predict ***' )
# Removing the `label` columns because it contains -1 and Trainer won't like that.
snake_case: Optional[int] = predict_dataset.remove_columns('label' )
snake_case: str = trainer.predict(__A , metric_key_prefix='predict' ).predictions
snake_case: Any = np.argmax(__A , axis=1 )
snake_case: int = os.path.join(training_args.output_dir , 'predict_results_tabfact.txt' )
if trainer.is_world_process_zero():
with open(__A , 'w' ) as writer:
logger.info('***** Predict Results *****' )
writer.write('index\tprediction\n' )
for index, item in enumerate(__A ):
snake_case: int = label_list[item]
writer.write(f"""{index}\t{item}\n""" )
snake_case: Optional[int] = {'finetuned_from': model_args.model_name_or_path, 'tasks': 'text-classification'}
if training_args.push_to_hub:
trainer.push_to_hub(**__A )
else:
trainer.create_model_card(**__A )
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
main()
if __name__ == "__main__":
main() | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
__UpperCAmelCase = [True] * 1_000_001
__UpperCAmelCase = 2
while i * i <= 1_000_000:
if seive[i]:
for j in range(i * i, 1_000_001, i):
__UpperCAmelCase = False
i += 1
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
return seive[n]
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
return any(digit in '02468' for digit in str(__A ) )
def lowerCAmelCase_ ( __A : int = 1_00_00_00 ):
'''simple docstring'''
snake_case: Optional[int] = [2] # result already includes the number 2.
for num in range(3 , limit + 1 , 2 ):
if is_prime(__A ) and not contains_an_even_digit(__A ):
snake_case: Optional[int] = str(__A )
snake_case: List[Any] = [int(str_num[j:] + str_num[:j] ) for j in range(len(__A ) )]
if all(is_prime(__A ) for i in list_nums ):
result.append(__A )
return result
def lowerCAmelCase_ ( ):
'''simple docstring'''
return len(find_circular_primes() )
if __name__ == "__main__":
print(F'{len(find_circular_primes()) = }') | 692 |
'''simple docstring'''
import math
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or number % 2 == 0 or number % 3 == 0:
# Negatives, 0, 1, all even numbers, all multiples of 3 are not primes
return False
# All primes number are in format of 6k +/- 1
for i in range(5 , int(math.sqrt(__A ) + 1 ) , 6 ):
if number % i == 0 or number % (i + 2) == 0:
return False
return True
def lowerCAmelCase_ ( __A : float = 0.1 ):
'''simple docstring'''
snake_case: Optional[int] = 3
snake_case: int = 3
while primes / (2 * j - 1) >= ratio:
for i in range(j * j + j + 1 , (j + 2) * (j + 2) , j + 1 ):
primes += is_prime(__A )
j += 2
return j
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ...utils.backbone_utils import BackboneConfigMixin, get_aligned_output_features_output_indices
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"facebook/convnextv2-tiny-1k-224": "https://huggingface.co/facebook/convnextv2-tiny-1k-224/resolve/main/config.json",
}
class SCREAMING_SNAKE_CASE ( snake_case , snake_case ):
'''simple docstring'''
__UpperCamelCase = "convnextv2"
def __init__( self , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-12 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: Dict = num_channels
snake_case: Optional[int] = patch_size
snake_case: List[str] = num_stages
snake_case: Dict = [96, 1_92, 3_84, 7_68] if hidden_sizes is None else hidden_sizes
snake_case: Union[str, Any] = [3, 3, 9, 3] if depths is None else depths
snake_case: Union[str, Any] = hidden_act
snake_case: List[Any] = initializer_range
snake_case: int = layer_norm_eps
snake_case: List[Any] = drop_path_rate
snake_case: Tuple = image_size
snake_case: Optional[int] = ['stem'] + [F"""stage{idx}""" for idx in range(1 , len(self.depths ) + 1 )]
snake_case , snake_case: int = get_aligned_output_features_output_indices(
out_features=SCREAMING_SNAKE_CASE__ , out_indices=SCREAMING_SNAKE_CASE__ , stage_names=self.stage_names ) | 692 |
'''simple docstring'''
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCAmelCase = "pt"
elif is_tf_available():
__UpperCAmelCase = "tf"
else:
__UpperCAmelCase = "jax"
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ByTaTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: int = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ByTaTokenizer.from_pretrained('google/byt5-small' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=20 , SCREAMING_SNAKE_CASE__=5 ):
'''simple docstring'''
snake_case: Optional[Any] = []
for i in range(len(SCREAMING_SNAKE_CASE__ ) ):
try:
snake_case: Optional[Any] = tokenizer.decode([i] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
except UnicodeDecodeError:
pass
toks.append((i, tok) )
snake_case: List[str] = list(filter(lambda SCREAMING_SNAKE_CASE__ : re.match(r'^[ a-zA-Z]+$' , t[1] ) , SCREAMING_SNAKE_CASE__ ) )
snake_case: str = list(filter(lambda SCREAMING_SNAKE_CASE__ : [t[0]] == tokenizer.encode(t[1] , add_special_tokens=SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ ) )
if max_length is not None and len(SCREAMING_SNAKE_CASE__ ) > max_length:
snake_case: Union[str, Any] = toks[:max_length]
if min_length is not None and len(SCREAMING_SNAKE_CASE__ ) < min_length and len(SCREAMING_SNAKE_CASE__ ) > 0:
while len(SCREAMING_SNAKE_CASE__ ) < min_length:
snake_case: Tuple = toks + toks
# toks_str = [t[1] for t in toks]
snake_case: Dict = [t[0] for t in toks]
# Ensure consistency
snake_case: int = tokenizer.decode(SCREAMING_SNAKE_CASE__ , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
if " " not in output_txt and len(SCREAMING_SNAKE_CASE__ ) > 1:
snake_case: str = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
)
if with_prefix_space:
snake_case: Tuple = ' ' + output_txt
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
return output_txt, output_ids
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: str = tokenizer(['hi</s>', 'I went to the gym</s>', '</s>'] )
snake_case: List[Any] = tokenizer(['hi', 'I went to the gym', ''] )
self.assertListEqual(batch_with_eos_added['input_ids'] , batch_without_eos_added['input_ids'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: Union[str, Any] = 'Unicode €.'
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [88, 1_13, 1_08, 1_02, 1_14, 1_03, 1_04, 35, 2_29, 1_33, 1_75, 49, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[str] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'Unicode €.</s>' )
snake_case: List[Any] = tokenizer('e è é ê ë' )
snake_case: Optional[Any] = [1_04, 35, 1_98, 1_71, 35, 1_98, 1_72, 35, 1_98, 1_73, 35, 1_98, 1_74, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[Any] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'e è é ê ë</s>' )
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë' ) ) , 'e è é ê ë</s>' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.ta_base_tokenizer
snake_case: Optional[Any] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case: Optional[int] = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 1, 0]
# fmt: on
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if FRAMEWORK != "jax":
snake_case: Optional[Any] = list(batch.input_ids.numpy()[0] )
else:
snake_case: Dict = list(batch.input_ids.tolist()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 37) , batch.input_ids.shape )
self.assertEqual((2, 37) , batch.attention_mask.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.ta_base_tokenizer
snake_case: List[str] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case: Union[str, Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_attention_mask' , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.ta_base_tokenizer
snake_case: str = [
'Summary of the text.',
'Another summary.',
]
snake_case: Dict = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.ta_base_tokenizer
snake_case: Optional[int] = ['A long paragraph for summarization. </s>']
snake_case: str = ['Summary of the text. </s>']
# fmt: off
snake_case: str = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 35, 1]
snake_case: Optional[int] = [86, 1_20, 1_12, 1_12, 1_00, 1_17, 1_24, 35, 1_14, 1_05, 35, 1_19, 1_07, 1_04, 35, 1_19, 1_04, 1_23, 1_19, 49, 35, 1]
# fmt: on
snake_case: List[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , text_target=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['input_ids'][0] )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['labels'][0] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: Optional[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: Union[str, Any] = tempfile.mkdtemp()
snake_case: Dict = ' He is very happy, UNwant\u00E9d,running'
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'] )
snake_case: List[str] = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token' )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: int = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn('new_additional_special_token' , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Union[str, Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: str = json.load(SCREAMING_SNAKE_CASE__ )
snake_case: int = [F"""<extra_id_{i}>""" for i in range(1_25 )]
snake_case: Optional[int] = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case: str = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Dict = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , )
self.assertIn(
'an_additional_special_token' , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['an_additional_special_token'] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'] ) ) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: Union[str, Any] = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token' , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: Union[str, Any] = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , )
self.assertIn('a_new_additional_special_token' , tokenizer.additional_special_tokens )
self.assertEqual(
['a_new_additional_special_token'] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'] ) ) , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertTrue(tokenizer.decode([2_55] ) == '' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_tokenizers(fast=SCREAMING_SNAKE_CASE__ , do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = ['t', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 'x', 't', '</s>']
snake_case: List[str] = tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Optional[Any] = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Dict = 0
snake_case: List[Any] = tokenizer.convert_ids_to_tokens(
SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [token_id_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [token_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [token_id_to_test_setters] ) | 692 | 1 |
'''simple docstring'''
from queue import Queue
from typing import TYPE_CHECKING, Optional
if TYPE_CHECKING:
from ..models.auto import AutoTokenizer
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
raise NotImplementedError()
def _UpperCamelCase ( self ):
'''simple docstring'''
raise NotImplementedError()
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = False , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = tokenizer
snake_case: Dict = skip_prompt
snake_case: Any = decode_kwargs
# variables used in the streaming process
snake_case: Tuple = []
snake_case: Tuple = 0
snake_case: List[str] = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if len(value.shape ) > 1 and value.shape[0] > 1:
raise ValueError('TextStreamer only supports batch size 1' )
elif len(value.shape ) > 1:
snake_case: Any = value[0]
if self.skip_prompt and self.next_tokens_are_prompt:
snake_case: str = False
return
# Add the new token to the cache and decodes the entire thing.
self.token_cache.extend(value.tolist() )
snake_case: int = self.tokenizer.decode(self.token_cache , **self.decode_kwargs )
# After the symbol for a new line, we flush the cache.
if text.endswith('\n' ):
snake_case: Union[str, Any] = text[self.print_len :]
snake_case: List[str] = []
snake_case: List[str] = 0
# If the last token is a CJK character, we print the characters.
elif len(SCREAMING_SNAKE_CASE__ ) > 0 and self._is_chinese_char(ord(text[-1] ) ):
snake_case: List[Any] = text[self.print_len :]
self.print_len += len(SCREAMING_SNAKE_CASE__ )
# Otherwise, prints until the last space char (simple heuristic to avoid printing incomplete words,
# which may change with the subsequent token -- there are probably smarter ways to do this!)
else:
snake_case: List[str] = text[self.print_len : text.rfind(' ' ) + 1]
self.print_len += len(SCREAMING_SNAKE_CASE__ )
self.on_finalized_text(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
if len(self.token_cache ) > 0:
snake_case: Any = self.tokenizer.decode(self.token_cache , **self.decode_kwargs )
snake_case: Optional[Any] = text[self.print_len :]
snake_case: str = []
snake_case: Tuple = 0
else:
snake_case: List[str] = ''
snake_case: Any = True
self.on_finalized_text(SCREAMING_SNAKE_CASE__ , stream_end=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
print(SCREAMING_SNAKE_CASE__ , flush=SCREAMING_SNAKE_CASE__ , end='' if not stream_end else None )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if (
(cp >= 0x4_e00 and cp <= 0x9_fff)
or (cp >= 0x3_400 and cp <= 0x4_dbf) #
or (cp >= 0x20_000 and cp <= 0x2a_6df) #
or (cp >= 0x2a_700 and cp <= 0x2b_73f) #
or (cp >= 0x2b_740 and cp <= 0x2b_81f) #
or (cp >= 0x2b_820 and cp <= 0x2c_eaf) #
or (cp >= 0xf_900 and cp <= 0xf_aff)
or (cp >= 0x2f_800 and cp <= 0x2f_a1f) #
): #
return True
return False
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = Queue()
snake_case: List[str] = None
snake_case: Tuple = timeout
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
self.text_queue.put(SCREAMING_SNAKE_CASE__ , timeout=self.timeout )
if stream_end:
self.text_queue.put(self.stop_signal , timeout=self.timeout )
def __iter__( self ):
'''simple docstring'''
return self
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.text_queue.get(timeout=self.timeout )
if value == self.stop_signal:
raise StopIteration()
else:
return value | 692 |
'''simple docstring'''
from typing import Any, Dict, Optional
import torch
import torch.nn.functional as F
from torch import nn
from ..utils import maybe_allow_in_graph
from .activations import get_activation
from .attention_processor import Attention
from .embeddings import CombinedTimestepLabelEmbeddings
@maybe_allow_in_graph
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "layer_norm" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = only_cross_attention
snake_case: Optional[Any] = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm_zero'
snake_case: Tuple = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm'
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
raise ValueError(
F"""`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"""
F""" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}.""" )
# Define 3 blocks. Each block has its own normalization layer.
# 1. Self-Attn
if self.use_ada_layer_norm:
snake_case: List[str] = AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case: str = AdaLayerNormZero(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
else:
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if only_cross_attention else None , upcast_attention=SCREAMING_SNAKE_CASE__ , )
# 2. Cross-Attn
if cross_attention_dim is not None or double_self_attention:
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
# the second cross attention block.
snake_case: Tuple = (
AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm
else nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if not double_self_attention else None , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , upcast_attention=SCREAMING_SNAKE_CASE__ , ) # is self-attn if encoder_hidden_states is none
else:
snake_case: int = None
snake_case: Tuple = None
# 3. Feed-forward
snake_case: Union[str, Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = FeedForward(SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , activation_fn=SCREAMING_SNAKE_CASE__ , final_dropout=SCREAMING_SNAKE_CASE__ )
# let chunk size default to None
snake_case: Any = None
snake_case: Any = 0
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = chunk_size
snake_case: str = dim
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
if self.use_ada_layer_norm:
snake_case: Optional[int] = self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case , snake_case , snake_case , snake_case , snake_case: int = self.norma(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=hidden_states.dtype )
else:
snake_case: List[str] = self.norma(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = cross_attention_kwargs if cross_attention_kwargs is not None else {}
snake_case: List[str] = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=encoder_hidden_states if self.only_cross_attention else None , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if self.use_ada_layer_norm_zero:
snake_case: Tuple = gate_msa.unsqueeze(1 ) * attn_output
snake_case: List[str] = attn_output + hidden_states
# 2. Cross-Attention
if self.attna is not None:
snake_case: Dict = (
self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) if self.use_ada_layer_norm else self.norma(SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: List[str] = attn_output + hidden_states
# 3. Feed-forward
snake_case: str = self.norma(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: str = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
if self._chunk_size is not None:
# "feed_forward_chunk_size" can be used to save memory
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
raise ValueError(
F"""`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`.""" )
snake_case: List[str] = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
snake_case: Optional[Any] = torch.cat(
[self.ff(SCREAMING_SNAKE_CASE__ ) for hid_slice in norm_hidden_states.chunk(SCREAMING_SNAKE_CASE__ , dim=self._chunk_dim )] , dim=self._chunk_dim , )
else:
snake_case: int = self.ff(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: Union[str, Any] = gate_mlp.unsqueeze(1 ) * ff_output
snake_case: Tuple = ff_output + hidden_states
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 4 , SCREAMING_SNAKE_CASE__ = 0.0 , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: int = int(dim * mult )
snake_case: Optional[Any] = dim_out if dim_out is not None else dim
if activation_fn == "gelu":
snake_case: int = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if activation_fn == "gelu-approximate":
snake_case: Optional[Any] = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , approximate='tanh' )
elif activation_fn == "geglu":
snake_case: List[Any] = GEGLU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif activation_fn == "geglu-approximate":
snake_case: Optional[int] = ApproximateGELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.ModuleList([] )
# project in
self.net.append(SCREAMING_SNAKE_CASE__ )
# project dropout
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
# project out
self.net.append(nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
if final_dropout:
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for module in self.net:
snake_case: Optional[int] = module(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "none" ):
'''simple docstring'''
super().__init__()
snake_case: Optional[int] = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = approximate
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ , approximate=self.approximate )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) , approximate=self.approximate ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.proj(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.gelu(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = nn.Linear(SCREAMING_SNAKE_CASE__ , dim_out * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: int = self.proj(SCREAMING_SNAKE_CASE__ ).chunk(2 , dim=-1 )
return hidden_states * self.gelu(SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = self.proj(SCREAMING_SNAKE_CASE__ )
return x * torch.sigmoid(1.7_02 * x )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Optional[Any] = nn.Embedding(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = nn.SiLU()
snake_case: Union[str, Any] = nn.Linear(SCREAMING_SNAKE_CASE__ , embedding_dim * 2 )
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case: Dict = torch.chunk(SCREAMING_SNAKE_CASE__ , 2 )
snake_case: str = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale) + shift
return x
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = CombinedTimestepLabelEmbeddings(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.SiLU()
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , 6 * embedding_dim , bias=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ , eps=1E-6 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
snake_case: int = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case , snake_case , snake_case , snake_case , snake_case: str = emb.chunk(6 , dim=1 )
snake_case: Dict = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale_msa[:, None]) + shift_msa[:, None]
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 1E-5 ):
'''simple docstring'''
super().__init__()
snake_case: str = num_groups
snake_case: str = eps
if act_fn is None:
snake_case: Dict = None
else:
snake_case: List[str] = get_activation(SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , out_dim * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.act:
snake_case: Optional[Any] = self.act(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.linear(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = emb[:, :, None, None]
snake_case , snake_case: List[Any] = emb.chunk(2 , dim=1 )
snake_case: Any = F.group_norm(SCREAMING_SNAKE_CASE__ , self.num_groups , eps=self.eps )
snake_case: Optional[int] = x * (1 + scale) + shift
return x | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
__UpperCAmelCase = {
"configuration_bridgetower": [
"BRIDGETOWER_PRETRAINED_CONFIG_ARCHIVE_MAP",
"BridgeTowerConfig",
"BridgeTowerTextConfig",
"BridgeTowerVisionConfig",
],
"processing_bridgetower": ["BridgeTowerProcessor"],
}
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["BridgeTowerImageProcessor"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"BRIDGETOWER_PRETRAINED_MODEL_ARCHIVE_LIST",
"BridgeTowerForContrastiveLearning",
"BridgeTowerForImageAndTextRetrieval",
"BridgeTowerForMaskedLM",
"BridgeTowerModel",
"BridgeTowerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_bridgetower import (
BRIDGETOWER_PRETRAINED_CONFIG_ARCHIVE_MAP,
BridgeTowerConfig,
BridgeTowerTextConfig,
BridgeTowerVisionConfig,
)
from .processing_bridgetower import BridgeTowerProcessor
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .image_processing_bridgetower import BridgeTowerImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_bridgetower import (
BRIDGETOWER_PRETRAINED_MODEL_ARCHIVE_LIST,
BridgeTowerForContrastiveLearning,
BridgeTowerForImageAndTextRetrieval,
BridgeTowerForMaskedLM,
BridgeTowerModel,
BridgeTowerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure) | 692 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
import argparse
import torch
from torch import nn
from transformers import MaMaaaConfig, MaMaaaForConditionalGeneration
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: str = [
'encoder.version',
'decoder.version',
'model.encoder.version',
'model.decoder.version',
'decoder.output_projection.weight',
'_float_tensor',
'encoder.embed_positions._float_tensor',
'decoder.embed_positions._float_tensor',
]
for k in ignore_keys:
state_dict.pop(__A , __A )
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
snake_case , snake_case: Optional[Any] = emb.weight.shape
snake_case: Optional[Any] = nn.Linear(__A , __A , bias=__A )
snake_case: List[str] = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
snake_case: Tuple = torch.load(__A , map_location='cpu' )
snake_case: Optional[int] = mam_aaa['args'] or mam_aaa['cfg']['model']
snake_case: Union[str, Any] = mam_aaa['model']
remove_ignore_keys_(__A )
snake_case: Optional[Any] = state_dict['encoder.embed_tokens.weight'].shape[0]
snake_case: str = MaMaaaConfig(
vocab_size=__A , max_position_embeddings=10_24 , encoder_layers=args.encoder_layers , decoder_layers=args.decoder_layers , encoder_attention_heads=args.encoder_attention_heads , decoder_attention_heads=args.decoder_attention_heads , encoder_ffn_dim=args.encoder_ffn_embed_dim , decoder_ffn_dim=args.decoder_ffn_embed_dim , d_model=args.encoder_embed_dim , encoder_layerdrop=args.encoder_layerdrop , decoder_layerdrop=args.decoder_layerdrop , dropout=args.dropout , attention_dropout=args.attention_dropout , activation_dropout=args.activation_dropout , activation_function='relu' , )
snake_case: Optional[int] = state_dict['decoder.embed_tokens.weight']
snake_case: Union[str, Any] = MaMaaaForConditionalGeneration(__A )
model.model.load_state_dict(__A , strict=__A )
snake_case: Optional[int] = make_linear_from_emb(model.model.shared )
return model
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument("fairseq_path", type=str, help="path to a model.pt on local filesystem.")
parser.add_argument("pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
__UpperCAmelCase = parser.parse_args()
__UpperCAmelCase = convert_fairseq_mamaaa_checkpoint_from_disk(args.fairseq_pathß)
model.save_pretrained(args.pytorch_dump_folder_path) | 692 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import convert_to_rgb, normalize, rescale, resize, to_channel_dimension_format
from ...image_utils import (
OPENAI_CLIP_MEAN,
OPENAI_CLIP_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
make_list_of_images,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = ["pixel_values"]
def __init__( self , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = PILImageResampling.BICUBIC , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = 1 / 2_55 , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = size if size is not None else {'height': 3_84, 'width': 3_84}
snake_case: int = get_size_dict(SCREAMING_SNAKE_CASE__ , default_to_square=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = do_resize
snake_case: int = size
snake_case: Optional[Any] = resample
snake_case: Optional[int] = do_rescale
snake_case: Tuple = rescale_factor
snake_case: Tuple = do_normalize
snake_case: Tuple = image_mean if image_mean is not None else OPENAI_CLIP_MEAN
snake_case: List[str] = image_std if image_std is not None else OPENAI_CLIP_STD
snake_case: Tuple = do_convert_rgb
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = PILImageResampling.BICUBIC , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: int = get_size_dict(SCREAMING_SNAKE_CASE__ , default_to_square=SCREAMING_SNAKE_CASE__ )
if "height" not in size or "width" not in size:
raise ValueError(F"""The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}""" )
snake_case: Tuple = (size['height'], size['width'])
return resize(SCREAMING_SNAKE_CASE__ , size=SCREAMING_SNAKE_CASE__ , resample=SCREAMING_SNAKE_CASE__ , data_format=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
return rescale(SCREAMING_SNAKE_CASE__ , scale=SCREAMING_SNAKE_CASE__ , data_format=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
return normalize(SCREAMING_SNAKE_CASE__ , mean=SCREAMING_SNAKE_CASE__ , std=SCREAMING_SNAKE_CASE__ , data_format=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = ChannelDimension.FIRST , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[int] = do_resize if do_resize is not None else self.do_resize
snake_case: Dict = resample if resample is not None else self.resample
snake_case: Optional[Any] = do_rescale if do_rescale is not None else self.do_rescale
snake_case: Optional[Any] = rescale_factor if rescale_factor is not None else self.rescale_factor
snake_case: Tuple = do_normalize if do_normalize is not None else self.do_normalize
snake_case: Any = image_mean if image_mean is not None else self.image_mean
snake_case: Optional[int] = image_std if image_std is not None else self.image_std
snake_case: int = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb
snake_case: Any = size if size is not None else self.size
snake_case: Optional[Any] = get_size_dict(SCREAMING_SNAKE_CASE__ , default_to_square=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = make_list_of_images(SCREAMING_SNAKE_CASE__ )
if not valid_images(SCREAMING_SNAKE_CASE__ ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# PIL RGBA images are converted to RGB
if do_convert_rgb:
snake_case: Any = [convert_to_rgb(SCREAMING_SNAKE_CASE__ ) for image in images]
# All transformations expect numpy arrays.
snake_case: Dict = [to_numpy_array(SCREAMING_SNAKE_CASE__ ) for image in images]
if do_resize:
snake_case: List[Any] = [self.resize(image=SCREAMING_SNAKE_CASE__ , size=SCREAMING_SNAKE_CASE__ , resample=SCREAMING_SNAKE_CASE__ ) for image in images]
if do_rescale:
snake_case: str = [self.rescale(image=SCREAMING_SNAKE_CASE__ , scale=SCREAMING_SNAKE_CASE__ ) for image in images]
if do_normalize:
snake_case: Any = [self.normalize(image=SCREAMING_SNAKE_CASE__ , mean=SCREAMING_SNAKE_CASE__ , std=SCREAMING_SNAKE_CASE__ ) for image in images]
snake_case: Any = [to_channel_dimension_format(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) for image in images]
snake_case: Optional[Any] = BatchFeature(data={'pixel_values': images} , tensor_type=SCREAMING_SNAKE_CASE__ )
return encoded_outputs | 692 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_roformer": ["ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoFormerConfig", "RoFormerOnnxConfig"],
"tokenization_roformer": ["RoFormerTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["RoFormerTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoFormerForCausalLM",
"RoFormerForMaskedLM",
"RoFormerForMultipleChoice",
"RoFormerForQuestionAnswering",
"RoFormerForSequenceClassification",
"RoFormerForTokenClassification",
"RoFormerLayer",
"RoFormerModel",
"RoFormerPreTrainedModel",
"load_tf_weights_in_roformer",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFRoFormerForCausalLM",
"TFRoFormerForMaskedLM",
"TFRoFormerForMultipleChoice",
"TFRoFormerForQuestionAnswering",
"TFRoFormerForSequenceClassification",
"TFRoFormerForTokenClassification",
"TFRoFormerLayer",
"TFRoFormerModel",
"TFRoFormerPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"FlaxRoFormerForMaskedLM",
"FlaxRoFormerForMultipleChoice",
"FlaxRoFormerForQuestionAnswering",
"FlaxRoFormerForSequenceClassification",
"FlaxRoFormerForTokenClassification",
"FlaxRoFormerModel",
"FlaxRoFormerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig
from .tokenization_roformer import RoFormerTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_roformer_fast import RoFormerTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roformer import (
ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
RoFormerForCausalLM,
RoFormerForMaskedLM,
RoFormerForMultipleChoice,
RoFormerForQuestionAnswering,
RoFormerForSequenceClassification,
RoFormerForTokenClassification,
RoFormerLayer,
RoFormerModel,
RoFormerPreTrainedModel,
load_tf_weights_in_roformer,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_roformer import (
TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TFRoFormerForCausalLM,
TFRoFormerForMaskedLM,
TFRoFormerForMultipleChoice,
TFRoFormerForQuestionAnswering,
TFRoFormerForSequenceClassification,
TFRoFormerForTokenClassification,
TFRoFormerLayer,
TFRoFormerModel,
TFRoFormerPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_roformer import (
FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
FlaxRoFormerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 | 1 |
'''simple docstring'''
import argparse
import json
from pathlib import Path
import requests
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import YolosConfig, YolosForObjectDetection, YolosImageProcessor
from transformers.utils import logging
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: Tuple = YolosConfig()
# size of the architecture
if "yolos_ti" in yolos_name:
snake_case: List[Any] = 1_92
snake_case: Dict = 7_68
snake_case: Dict = 12
snake_case: List[str] = 3
snake_case: Any = [8_00, 13_33]
snake_case: Any = False
elif yolos_name == "yolos_s_dWr":
snake_case: int = 3_30
snake_case: Tuple = 14
snake_case: Tuple = 6
snake_case: Dict = 13_20
elif "yolos_s" in yolos_name:
snake_case: str = 3_84
snake_case: str = 15_36
snake_case: Any = 12
snake_case: int = 6
elif "yolos_b" in yolos_name:
snake_case: List[Any] = [8_00, 13_44]
snake_case: Union[str, Any] = 91
snake_case: List[str] = 'huggingface/label-files'
snake_case: Optional[Any] = 'coco-detection-id2label.json'
snake_case: Dict = json.load(open(hf_hub_download(__A , __A , repo_type='dataset' ) , 'r' ) )
snake_case: Any = {int(__A ): v for k, v in idalabel.items()}
snake_case: Any = idalabel
snake_case: Union[str, Any] = {v: k for k, v in idalabel.items()}
return config
def lowerCAmelCase_ ( __A : dict , __A : YolosConfig , __A : bool = False ):
'''simple docstring'''
for i in range(config.num_hidden_layers ):
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case: Optional[Any] = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case: Union[str, Any] = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case: Optional[Any] = in_proj_weight[: config.hidden_size, :]
snake_case: Optional[int] = in_proj_bias[: config.hidden_size]
snake_case: Tuple = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case: List[Any] = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case: str = in_proj_weight[-config.hidden_size :, :]
snake_case: List[str] = in_proj_bias[-config.hidden_size :]
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
if "backbone" in name:
snake_case: List[str] = name.replace('backbone' , 'vit' )
if "cls_token" in name:
snake_case: Optional[int] = name.replace('cls_token' , 'embeddings.cls_token' )
if "det_token" in name:
snake_case: Tuple = name.replace('det_token' , 'embeddings.detection_tokens' )
if "mid_pos_embed" in name:
snake_case: List[Any] = name.replace('mid_pos_embed' , 'encoder.mid_position_embeddings' )
if "pos_embed" in name:
snake_case: Optional[int] = name.replace('pos_embed' , 'embeddings.position_embeddings' )
if "patch_embed.proj" in name:
snake_case: Dict = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "blocks" in name:
snake_case: Tuple = name.replace('blocks' , 'encoder.layer' )
if "attn.proj" in name:
snake_case: Optional[Any] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name:
snake_case: Optional[int] = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Any = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: Optional[Any] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Any = name.replace('mlp.fc2' , 'output.dense' )
if "class_embed" in name:
snake_case: List[Any] = name.replace('class_embed' , 'class_labels_classifier' )
if "bbox_embed" in name:
snake_case: str = name.replace('bbox_embed' , 'bbox_predictor' )
if "vit.norm" in name:
snake_case: List[Any] = name.replace('vit.norm' , 'vit.layernorm' )
return name
def lowerCAmelCase_ ( __A : dict , __A : YolosForObjectDetection ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: Union[str, Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Dict = key.split('.' )
snake_case: List[Any] = int(key_split[2] )
snake_case: List[str] = model.vit.encoder.layer[layer_num].attention.attention.all_head_size
if "weight" in key:
snake_case: Tuple = val[:dim, :]
snake_case: Any = val[
dim : dim * 2, :
]
snake_case: Tuple = val[-dim:, :]
else:
snake_case: int = val[:dim]
snake_case: str = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
else:
snake_case: List[str] = val
return orig_state_dict
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'http://images.cocodataset.org/val2017/000000039769.jpg'
snake_case: List[Any] = Image.open(requests.get(__A , stream=__A ).raw )
return im
@torch.no_grad()
def lowerCAmelCase_ ( __A : str , __A : str , __A : str , __A : bool = False ):
'''simple docstring'''
snake_case: str = get_yolos_config(__A )
# load original state_dict
snake_case: List[Any] = torch.load(__A , map_location='cpu' )['model']
# load 🤗 model
snake_case: Dict = YolosForObjectDetection(__A )
model.eval()
snake_case: str = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# Check outputs on an image, prepared by YolosImageProcessor
snake_case: Tuple = 8_00 if yolos_name != 'yolos_ti' else 5_12
snake_case: Any = YolosImageProcessor(format='coco_detection' , size=__A )
snake_case: Union[str, Any] = image_processor(images=prepare_img() , return_tensors='pt' )
snake_case: Any = model(**__A )
snake_case , snake_case: Optional[Any] = outputs.logits, outputs.pred_boxes
snake_case , snake_case: Optional[int] = None, None
if yolos_name == "yolos_ti":
snake_case: Optional[Any] = torch.tensor(
[[-39.50_22, -11.98_20, -17.68_88], [-29.95_74, -9.97_69, -17.76_91], [-42.32_81, -20.72_00, -30.62_94]] )
snake_case: Optional[int] = torch.tensor(
[[0.40_21, 0.08_36, 0.79_79], [0.01_84, 0.26_09, 0.03_64], [0.17_81, 0.20_04, 0.20_95]] )
elif yolos_name == "yolos_s_200_pre":
snake_case: Dict = torch.tensor(
[[-24.02_48, -10.30_24, -14.82_90], [-42.03_92, -16.82_00, -27.43_34], [-27.27_43, -11.81_54, -18.71_48]] )
snake_case: List[Any] = torch.tensor(
[[0.25_59, 0.54_55, 0.47_06], [0.29_89, 0.72_79, 0.18_75], [0.77_32, 0.40_17, 0.44_62]] )
elif yolos_name == "yolos_s_300_pre":
snake_case: int = torch.tensor(
[[-36.22_20, -14.43_85, -23.54_57], [-35.69_70, -14.75_83, -21.39_35], [-31.59_39, -13.60_42, -16.80_49]] )
snake_case: List[Any] = torch.tensor(
[[0.76_14, 0.23_16, 0.47_28], [0.71_68, 0.44_95, 0.38_55], [0.49_96, 0.14_66, 0.99_96]] )
elif yolos_name == "yolos_s_dWr":
snake_case: Tuple = torch.tensor(
[[-42.86_68, -24.10_49, -41.16_90], [-34.74_56, -14.12_74, -24.91_94], [-33.78_98, -12.19_46, -25.64_95]] )
snake_case: Any = torch.tensor(
[[0.55_87, 0.27_73, 0.06_05], [0.50_04, 0.30_14, 0.99_94], [0.49_99, 0.15_48, 0.99_94]] )
elif yolos_name == "yolos_base":
snake_case: str = torch.tensor(
[[-40.60_64, -24.30_84, -32.64_47], [-55.19_90, -30.77_19, -35.58_77], [-51.43_11, -33.35_07, -35.64_62]] )
snake_case: Dict = torch.tensor(
[[0.55_55, 0.27_94, 0.06_55], [0.90_49, 0.26_64, 0.18_94], [0.91_83, 0.19_84, 0.16_35]] )
else:
raise ValueError(f"""Unknown yolos_name: {yolos_name}""" )
assert torch.allclose(logits[0, :3, :3] , __A , atol=1E-4 )
assert torch.allclose(pred_boxes[0, :3, :3] , __A , atol=1E-4 )
Path(__A ).mkdir(exist_ok=__A )
print(f"""Saving model {yolos_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(__A )
if push_to_hub:
snake_case: Any = {
'yolos_ti': 'yolos-tiny',
'yolos_s_200_pre': 'yolos-small',
'yolos_s_300_pre': 'yolos-small-300',
'yolos_s_dWr': 'yolos-small-dwr',
'yolos_base': 'yolos-base',
}
print('Pushing to the hub...' )
snake_case: int = model_mapping[yolos_name]
image_processor.push_to_hub(__A , organization='hustvl' )
model.push_to_hub(__A , organization='hustvl' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--yolos_name",
default="yolos_s_200_pre",
type=str,
help=(
"Name of the YOLOS model you'd like to convert. Should be one of 'yolos_ti', 'yolos_s_200_pre',"
" 'yolos_s_300_pre', 'yolos_s_dWr', 'yolos_base'."
),
)
parser.add_argument(
"--checkpoint_path", default=None, type=str, help="Path to the original state dict (.pth file)."
)
parser.add_argument(
"--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model directory."
)
parser.add_argument(
"--push_to_hub", action="store_true", help="Whether or not to push the converted model to the 🤗 hub."
)
__UpperCAmelCase = parser.parse_args()
convert_yolos_checkpoint(args.yolos_name, args.checkpoint_path, args.pytorch_dump_folder_path, args.push_to_hub) | 692 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
from random import random
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: Optional[Any] = value
snake_case: List[str] = random()
snake_case: Node | None = None
snake_case: Node | None = None
def __repr__( self ):
'''simple docstring'''
from pprint import pformat
if self.left is None and self.right is None:
return F"""'{self.value}: {self.prior:.5}'"""
else:
return pformat(
{F"""{self.value}: {self.prior:.5}""": (self.left, self.right)} , indent=1 )
def __str__( self ):
'''simple docstring'''
snake_case: Union[str, Any] = str(self.value ) + ' '
snake_case: str = str(self.left or '' )
snake_case: Union[str, Any] = str(self.right or '' )
return value + left + right
def lowerCAmelCase_ ( __A : Node | None , __A : int ):
'''simple docstring'''
if root is None: # None tree is split into 2 Nones
return None, None
elif root.value is None:
return None, None
else:
if value < root.value:
snake_case , snake_case: Dict = split(root.left , __A )
return left, root
else:
snake_case , snake_case: Any = split(root.right , __A )
return root, right
def lowerCAmelCase_ ( __A : Node | None , __A : Node | None ):
'''simple docstring'''
if (not left) or (not right): # If one node is None, return the other
return left or right
elif left.prior < right.prior:
snake_case: Optional[Any] = merge(left.right , __A )
return left
else:
snake_case: List[Any] = merge(__A , right.left )
return right
def lowerCAmelCase_ ( __A : Node | None , __A : int ):
'''simple docstring'''
snake_case: Optional[Any] = Node(__A )
snake_case , snake_case: int = split(__A , __A )
return merge(merge(__A , __A ) , __A )
def lowerCAmelCase_ ( __A : Node | None , __A : int ):
'''simple docstring'''
snake_case , snake_case: List[Any] = split(__A , value - 1 )
snake_case , snake_case: Any = split(__A , __A )
return merge(__A , __A )
def lowerCAmelCase_ ( __A : Node | None ):
'''simple docstring'''
if not root: # None
return
else:
inorder(root.left )
print(root.value , end=',' )
inorder(root.right )
def lowerCAmelCase_ ( __A : Node | None , __A : str ):
'''simple docstring'''
for arg in args.split():
if arg[0] == "+":
snake_case: str = insert(__A , int(arg[1:] ) )
elif arg[0] == "-":
snake_case: Dict = erase(__A , int(arg[1:] ) )
else:
print('Unknown command' )
return root
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Optional[int] = None
print(
'enter numbers to create a tree, + value to add value into treap, '
'- value to erase all nodes with value. \'q\' to quit. ' )
snake_case: Any = input()
while args != "q":
snake_case: str = interact_treap(__A , __A )
print(__A )
snake_case: List[Any] = input()
print('good by!' )
if __name__ == "__main__":
import doctest
doctest.testmod()
main() | 692 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 | 1 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int , __A : int , __A : list[list[int]] ):
'''simple docstring'''
def update_area_of_max_square(__A : int , __A : int ) -> int:
# BASE CASE
if row >= rows or col >= cols:
return 0
snake_case: List[Any] = update_area_of_max_square(__A , col + 1 )
snake_case: Dict = update_area_of_max_square(row + 1 , col + 1 )
snake_case: List[Any] = update_area_of_max_square(row + 1 , __A )
if mat[row][col]:
snake_case: Optional[int] = 1 + min([right, diagonal, down] )
snake_case: int = max(largest_square_area[0] , __A )
return sub_problem_sol
else:
return 0
snake_case: Tuple = [0]
update_area_of_max_square(0 , 0 )
return largest_square_area[0]
def lowerCAmelCase_ ( __A : int , __A : int , __A : list[list[int]] ):
'''simple docstring'''
def update_area_of_max_square_using_dp_array(
__A : int , __A : int , __A : list[list[int]] ) -> int:
if row >= rows or col >= cols:
return 0
if dp_array[row][col] != -1:
return dp_array[row][col]
snake_case: Optional[Any] = update_area_of_max_square_using_dp_array(__A , col + 1 , __A )
snake_case: Dict = update_area_of_max_square_using_dp_array(row + 1 , col + 1 , __A )
snake_case: List[Any] = update_area_of_max_square_using_dp_array(row + 1 , __A , __A )
if mat[row][col]:
snake_case: str = 1 + min([right, diagonal, down] )
snake_case: Any = max(largest_square_area[0] , __A )
snake_case: int = sub_problem_sol
return sub_problem_sol
else:
return 0
snake_case: Tuple = [0]
snake_case: Union[str, Any] = [[-1] * cols for _ in range(__A )]
update_area_of_max_square_using_dp_array(0 , 0 , __A )
return largest_square_area[0]
def lowerCAmelCase_ ( __A : int , __A : int , __A : list[list[int]] ):
'''simple docstring'''
snake_case: str = [[0] * (cols + 1) for _ in range(rows + 1 )]
snake_case: Dict = 0
for row in range(rows - 1 , -1 , -1 ):
for col in range(cols - 1 , -1 , -1 ):
snake_case: Optional[int] = dp_array[row][col + 1]
snake_case: Optional[int] = dp_array[row + 1][col + 1]
snake_case: int = dp_array[row + 1][col]
if mat[row][col] == 1:
snake_case: Optional[Any] = 1 + min(__A , __A , __A )
snake_case: List[Any] = max(dp_array[row][col] , __A )
else:
snake_case: Optional[Any] = 0
return largest_square_area
def lowerCAmelCase_ ( __A : int , __A : int , __A : list[list[int]] ):
'''simple docstring'''
snake_case: str = [0] * (cols + 1)
snake_case: Union[str, Any] = [0] * (cols + 1)
snake_case: Union[str, Any] = 0
for row in range(rows - 1 , -1 , -1 ):
for col in range(cols - 1 , -1 , -1 ):
snake_case: Any = current_row[col + 1]
snake_case: str = next_row[col + 1]
snake_case: List[str] = next_row[col]
if mat[row][col] == 1:
snake_case: Dict = 1 + min(__A , __A , __A )
snake_case: Dict = max(current_row[col] , __A )
else:
snake_case: str = 0
snake_case: List[str] = current_row
return largest_square_area
if __name__ == "__main__":
import doctest
doctest.testmod()
print(largest_square_area_in_matrix_bottom_up(2, 2, [[1, 1], [1, 1]])) | 692 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
__UpperCAmelCase = {
"configuration_instructblip": [
"INSTRUCTBLIP_PRETRAINED_CONFIG_ARCHIVE_MAP",
"InstructBlipConfig",
"InstructBlipQFormerConfig",
"InstructBlipVisionConfig",
],
"processing_instructblip": ["InstructBlipProcessor"],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"INSTRUCTBLIP_PRETRAINED_MODEL_ARCHIVE_LIST",
"InstructBlipQFormerModel",
"InstructBlipPreTrainedModel",
"InstructBlipForConditionalGeneration",
"InstructBlipVisionModel",
]
if TYPE_CHECKING:
from .configuration_instructblip import (
INSTRUCTBLIP_PRETRAINED_CONFIG_ARCHIVE_MAP,
InstructBlipConfig,
InstructBlipQFormerConfig,
InstructBlipVisionConfig,
)
from .processing_instructblip import InstructBlipProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_instructblip import (
INSTRUCTBLIP_PRETRAINED_MODEL_ARCHIVE_LIST,
InstructBlipForConditionalGeneration,
InstructBlipPreTrainedModel,
InstructBlipQFormerModel,
InstructBlipVisionModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = "▁"
__UpperCAmelCase = {"vocab_file": "sentencepiece.bpe.model"}
__UpperCAmelCase = {
"vocab_file": {
"facebook/xglm-564M": "https://huggingface.co/facebook/xglm-564M/resolve/main/sentencepiece.bpe.model",
}
}
__UpperCAmelCase = {
"facebook/xglm-564M": 2_048,
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
# Compatibility with the original tokenizer
snake_case: Optional[Any] = 7
snake_case: List[str] = [F"""<madeupword{i}>""" for i in range(self.num_madeup_words )]
snake_case: str = kwargs.get('additional_special_tokens' , [] )
kwargs["additional_special_tokens"] += [
word for word in madeup_words if word not in kwargs["additional_special_tokens"]
]
super().__init__(
bos_token=SCREAMING_SNAKE_CASE__ , eos_token=SCREAMING_SNAKE_CASE__ , unk_token=SCREAMING_SNAKE_CASE__ , sep_token=SCREAMING_SNAKE_CASE__ , cls_token=SCREAMING_SNAKE_CASE__ , pad_token=SCREAMING_SNAKE_CASE__ , sp_model_kwargs=self.sp_model_kwargs , **SCREAMING_SNAKE_CASE__ , )
snake_case: int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(SCREAMING_SNAKE_CASE__ ) )
snake_case: int = vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
snake_case: Tuple = 1
# Mimic fairseq token-to-id alignment for the first 4 token
snake_case: Optional[Any] = {'<s>': 0, '<pad>': 1, '</s>': 2, '<unk>': 3}
snake_case: Union[str, Any] = len(self.sp_model )
snake_case: str = {F"""<madeupword{i}>""": sp_size + i + self.fairseq_offset for i in range(self.num_madeup_words )}
self.fairseq_tokens_to_ids.update(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def __getstate__( self ):
'''simple docstring'''
snake_case: List[Any] = self.__dict__.copy()
snake_case: Union[str, Any] = None
snake_case: Union[str, Any] = self.sp_model.serialized_model_proto()
return state
def __setstate__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
snake_case: Union[str, Any] = {}
snake_case: Tuple = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.LoadFromSerializedProto(self.sp_model_proto )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.sep_token_id] + token_ids_a
snake_case: Optional[Any] = [self.sep_token_id]
return sep + token_ids_a + sep + sep + token_ids_a
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=SCREAMING_SNAKE_CASE__ , token_ids_a=SCREAMING_SNAKE_CASE__ , already_has_special_tokens=SCREAMING_SNAKE_CASE__ )
if token_ids_a is None:
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ )) + [1, 1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
if token_ids_a is None:
return len(sep + token_ids_a ) * [0]
return len(sep + token_ids_a + sep + sep + token_ids_a ) * [0]
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.sp_model ) + self.fairseq_offset + self.num_madeup_words
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = {self.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.sp_model.encode(SCREAMING_SNAKE_CASE__ , out_type=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
snake_case: Dict = self.sp_model.PieceToId(SCREAMING_SNAKE_CASE__ )
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = ''.join(SCREAMING_SNAKE_CASE__ ).replace(SCREAMING_SNAKE_CASE__ , ' ' ).strip()
return out_string
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(SCREAMING_SNAKE_CASE__ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: List[str] = os.path.join(
SCREAMING_SNAKE_CASE__ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(SCREAMING_SNAKE_CASE__ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , SCREAMING_SNAKE_CASE__ )
elif not os.path.isfile(self.vocab_file ):
with open(SCREAMING_SNAKE_CASE__ , 'wb' ) as fi:
snake_case: int = self.sp_model.serialized_model_proto()
fi.write(SCREAMING_SNAKE_CASE__ )
return (out_vocab_file,) | 692 | 1 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"facebook/timesformer": "https://huggingface.co/facebook/timesformer/resolve/main/config.json",
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "timesformer"
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=8 , SCREAMING_SNAKE_CASE__=7_68 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=30_72 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-6 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__="divided_space_time" , SCREAMING_SNAKE_CASE__=0 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = image_size
snake_case: Tuple = patch_size
snake_case: int = num_channels
snake_case: int = num_frames
snake_case: Tuple = hidden_size
snake_case: Optional[int] = num_hidden_layers
snake_case: Any = num_attention_heads
snake_case: Tuple = intermediate_size
snake_case: List[str] = hidden_act
snake_case: int = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: List[str] = initializer_range
snake_case: int = layer_norm_eps
snake_case: str = qkv_bias
snake_case: Optional[int] = attention_type
snake_case: List[str] = drop_path_rate | 692 |
'''simple docstring'''
from operator import delitem, getitem, setitem
import pytest
from data_structures.hashing.hash_map import HashMap
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
return getitem, k
def lowerCAmelCase_ ( __A : Any , __A : Optional[int] ):
'''simple docstring'''
return setitem, k, v
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
return delitem, k
def lowerCAmelCase_ ( __A : str , __A : int , *__A : Tuple ):
'''simple docstring'''
try:
return fun(__A , *__A ), None
except Exception as e:
return None, e
__UpperCAmelCase = (
_set("key_a", "val_a"),
_set("key_b", "val_b"),
)
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_a", "val_b"),
]
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_b", "val_b"),
_del("key_a"),
_del("key_b"),
_set("key_a", "val_a"),
_del("key_a"),
]
__UpperCAmelCase = [
_get("key_a"),
_del("key_a"),
_set("key_a", "val_a"),
_del("key_a"),
_del("key_a"),
_get("key_a"),
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
*[_del(x) for x in range(5)],
_set("key_a", "val_b"),
]
@pytest.mark.parametrize(
'operations' , (
pytest.param(_add_items , id='add items' ),
pytest.param(_overwrite_items , id='overwrite items' ),
pytest.param(_delete_items , id='delete items' ),
pytest.param(_access_absent_items , id='access absent items' ),
pytest.param(_add_with_resize_up , id='add with resize up' ),
pytest.param(_add_with_resize_down , id='add with resize down' ),
) , )
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[Any] = HashMap(initial_block_size=4 )
snake_case: List[Any] = {}
for _, (fun, *args) in enumerate(__A ):
snake_case , snake_case: Optional[int] = _run_operation(__A , __A , *__A )
snake_case , snake_case: str = _run_operation(__A , __A , *__A )
assert my_res == py_res
assert str(__A ) == str(__A )
assert set(__A ) == set(__A )
assert len(__A ) == len(__A )
assert set(my.items() ) == set(py.items() )
def lowerCAmelCase_ ( ):
'''simple docstring'''
def is_public(__A : str ) -> bool:
return not name.startswith('_' )
snake_case: Dict = {name for name in dir({} ) if is_public(__A )}
snake_case: List[str] = {name for name in dir(HashMap() ) if is_public(__A )}
assert dict_public_names > hash_public_names | 692 | 1 |
'''simple docstring'''
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"EleutherAI/gpt-j-6B": "https://huggingface.co/EleutherAI/gpt-j-6B/resolve/main/config.json",
# See all GPT-J models at https://huggingface.co/models?filter=gpt_j
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "gptj"
__UpperCamelCase = {
"max_position_embeddings": "n_positions",
"hidden_size": "n_embd",
"num_attention_heads": "n_head",
"num_hidden_layers": "n_layer",
}
def __init__( self , SCREAMING_SNAKE_CASE__=5_04_00 , SCREAMING_SNAKE_CASE__=20_48 , SCREAMING_SNAKE_CASE__=40_96 , SCREAMING_SNAKE_CASE__=28 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=64 , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__="gelu_new" , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=5_02_56 , SCREAMING_SNAKE_CASE__=5_02_56 , SCREAMING_SNAKE_CASE__=False , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[int] = vocab_size
snake_case: Any = n_positions
snake_case: Tuple = n_embd
snake_case: Tuple = n_layer
snake_case: List[str] = n_head
snake_case: Optional[Any] = n_inner
snake_case: Optional[Any] = rotary_dim
snake_case: Dict = activation_function
snake_case: int = resid_pdrop
snake_case: List[Any] = embd_pdrop
snake_case: Optional[Any] = attn_pdrop
snake_case: Optional[Any] = layer_norm_epsilon
snake_case: Union[str, Any] = initializer_range
snake_case: Any = use_cache
snake_case: Optional[int] = bos_token_id
snake_case: List[str] = eos_token_id
super().__init__(
bos_token_id=SCREAMING_SNAKE_CASE__ , eos_token_id=SCREAMING_SNAKE_CASE__ , tie_word_embeddings=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "default" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ , task=SCREAMING_SNAKE_CASE__ , patching_specs=SCREAMING_SNAKE_CASE__ , use_past=SCREAMING_SNAKE_CASE__ )
if not getattr(self._config , 'pad_token_id' , SCREAMING_SNAKE_CASE__ ):
# TODO: how to do that better?
snake_case: Optional[Any] = 0
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = OrderedDict({'input_ids': {0: 'batch', 1: 'sequence'}} )
if self.use_past:
self.fill_with_past_key_values_(SCREAMING_SNAKE_CASE__ , direction='inputs' )
snake_case: Any = {0: 'batch', 1: 'past_sequence + sequence'}
else:
snake_case: str = {0: 'batch', 1: 'sequence'}
return common_inputs
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return self._config.n_layer
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return self._config.n_head
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = -1 , SCREAMING_SNAKE_CASE__ = -1 , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = super(SCREAMING_SNAKE_CASE__ , self ).generate_dummy_inputs(
SCREAMING_SNAKE_CASE__ , batch_size=SCREAMING_SNAKE_CASE__ , seq_length=SCREAMING_SNAKE_CASE__ , is_pair=SCREAMING_SNAKE_CASE__ , framework=SCREAMING_SNAKE_CASE__ )
# We need to order the input in the way they appears in the forward()
snake_case: Optional[int] = OrderedDict({'input_ids': common_inputs['input_ids']} )
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('Cannot generate dummy past_keys inputs without PyTorch installed.' )
else:
import torch
snake_case , snake_case: int = common_inputs['input_ids'].shape
# Not using the same length for past_key_values
snake_case: Any = seqlen + 2
snake_case: Any = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
snake_case: Dict = [
(torch.zeros(SCREAMING_SNAKE_CASE__ ), torch.zeros(SCREAMING_SNAKE_CASE__ )) for _ in range(self.num_layers )
]
snake_case: Optional[int] = common_inputs['attention_mask']
if self.use_past:
snake_case: List[Any] = ordered_inputs['attention_mask'].dtype
snake_case: str = torch.cat(
[ordered_inputs['attention_mask'], torch.ones(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , dtype=SCREAMING_SNAKE_CASE__ )] , dim=1 )
return ordered_inputs
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 13 | 692 |
'''simple docstring'''
import argparse
import json
import os
import fairseq
import torch
from torch import nn
from transformers import (
SpeechaTextaConfig,
SpeechaTextaForCausalLM,
SpeechaTextaTokenizer,
SpeechEncoderDecoderConfig,
SpeechEncoderDecoderModel,
WavaVecaConfig,
WavaVecaFeatureExtractor,
WavaVecaModel,
logging,
)
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"post_extract_proj": "feature_projection.projection",
"encoder.pos_conv.0": "encoder.pos_conv_embed.conv",
"self_attn.k_proj": "encoder.layers.*.attention.k_proj",
"self_attn.v_proj": "encoder.layers.*.attention.v_proj",
"self_attn.q_proj": "encoder.layers.*.attention.q_proj",
"self_attn.out_proj": "encoder.layers.*.attention.out_proj",
"self_attn_layer_norm": "encoder.layers.*.layer_norm",
"fc1": "encoder.layers.*.feed_forward.intermediate_dense",
"fc2": "encoder.layers.*.feed_forward.output_dense",
"final_layer_norm": "encoder.layers.*.final_layer_norm",
"encoder.layer_norm": "encoder.layer_norm",
"w2v_model.layer_norm": "feature_projection.layer_norm",
"quantizer.weight_proj": "quantizer.weight_proj",
"quantizer.vars": "quantizer.codevectors",
"project_q": "project_q",
"final_proj": "project_hid",
"w2v_encoder.proj": "lm_head",
"mask_emb": "masked_spec_embed",
}
__UpperCAmelCase = [
"lm_head",
"quantizer.weight_proj",
"quantizer.codevectors",
"project_q",
"project_hid",
]
def lowerCAmelCase_ ( __A : Any , __A : Optional[Any] , __A : Union[str, Any] , __A : int , __A : Optional[int] ):
'''simple docstring'''
for attribute in key.split('.' ):
snake_case: List[str] = getattr(__A , __A )
if weight_type is not None:
snake_case: Optional[int] = getattr(__A , __A ).shape
else:
snake_case: Optional[int] = hf_pointer.shape
assert hf_shape == value.shape, (
f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
snake_case: Optional[int] = value
elif weight_type == "weight_g":
snake_case: List[str] = value
elif weight_type == "weight_v":
snake_case: Dict = value
elif weight_type == "bias":
snake_case: Optional[Any] = value
else:
snake_case: int = value
logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def lowerCAmelCase_ ( __A : List[Any] , __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = []
snake_case: List[Any] = fairseq_model.state_dict()
snake_case: Union[str, Any] = hf_model.feature_extractor
# if encoder has different dim to decoder -> use proj_weight
snake_case: Dict = None
for name, value in fairseq_dict.items():
snake_case: Tuple = False
if "conv_layers" in name:
load_conv_layer(
__A , __A , __A , __A , hf_model.config.feat_extract_norm == 'group' , )
snake_case: List[Any] = True
elif name.split('.' )[0] == "proj":
snake_case: List[Any] = fairseq_model.proj
snake_case: int = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
snake_case: int = True
if "*" in mapped_key:
snake_case: List[str] = name.split(__A )[0].split('.' )[-2]
snake_case: Dict = mapped_key.replace('*' , __A )
if "weight_g" in name:
snake_case: Tuple = 'weight_g'
elif "weight_v" in name:
snake_case: int = 'weight_v'
elif "bias" in name:
snake_case: Tuple = 'bias'
elif "weight" in name:
snake_case: List[Any] = 'weight'
else:
snake_case: Any = None
set_recursively(__A , __A , __A , __A , __A )
continue
if not is_used:
unused_weights.append(__A )
logger.warning(f"""Unused weights: {unused_weights}""" )
return proj_weight
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : int , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = full_name.split('conv_layers.' )[-1]
snake_case: Tuple = name.split('.' )
snake_case: Any = int(items[0] )
snake_case: Optional[int] = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
snake_case: Tuple = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
snake_case: int = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
snake_case: Any = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
snake_case: str = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(__A )
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case , snake_case: List[Any] = emb.weight.shape
snake_case: Optional[int] = nn.Linear(__A , __A , bias=__A )
snake_case: Any = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
with open(__A , 'r' , encoding='utf-8' ) as f:
snake_case: List[Any] = f.readlines()
snake_case: Any = [line.split(' ' )[0] for line in lines]
snake_case: int = len(__A )
snake_case: Dict = {
'<s>': 0,
'<pad>': 1,
'</s>': 2,
'<unk>': 3,
}
vocab_dict.update(dict(zip(__A , range(4 , num_words + 4 ) ) ) )
return vocab_dict
@torch.no_grad()
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Dict , __A : Any , __A : List[Any] , __A : int , __A : str , ):
'''simple docstring'''
snake_case: Union[str, Any] = WavaVecaConfig.from_pretrained(__A )
snake_case: str = SpeechaTextaConfig.from_pretrained(
__A , vocab_size=__A , decoder_layers=__A , do_stable_layer_norm=__A )
snake_case: List[str] = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=1_60_00 , padding_value=0 , do_normalize=__A , return_attention_mask=__A , )
snake_case , snake_case , snake_case: List[str] = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
snake_case: List[Any] = model[0].eval()
# set weights for wav2vec2 encoder
snake_case: Optional[Any] = WavaVecaModel(__A )
snake_case: Any = recursively_load_weights_wavaveca(model.encoder , __A )
snake_case: Union[str, Any] = SpeechaTextaForCausalLM(__A )
snake_case , snake_case: Optional[Any] = hf_decoder.model.decoder.load_state_dict(model.decoder.state_dict() , strict=__A )
# set output linear layer
unexpected_keys.remove('embed_out' )
snake_case: str = nn.Parameter(model.decoder.embed_out.detach() )
# layer norm is init to identity matrix so leaving it is fine
logger.warning(f"""The following keys are missing when loading the decoder weights: {missing_keys}""" )
logger.warning(f"""The following keys are unexpected when loading the decoder weights: {unexpected_keys}""" )
snake_case: int = SpeechEncoderDecoderModel(encoder=__A , decoder=__A )
snake_case: List[Any] = False
# add projection layer
snake_case: Union[str, Any] = nn.Parameter(projection_layer.weight )
snake_case: Union[str, Any] = nn.Parameter(projection_layer.bias )
snake_case: List[Any] = create_vocab_dict(__A )
with open(os.path.join(__A , 'vocab.json' ) , 'w' ) as fp:
json.dump(__A , __A )
snake_case: Union[str, Any] = SpeechaTextaTokenizer(os.path.join(__A , 'vocab.json' ) )
tokenizer.save_pretrained(__A )
snake_case: Tuple = hf_wavavec.config.to_dict()
snake_case: int = tokenizer.pad_token_id
snake_case: Dict = tokenizer.bos_token_id
snake_case: Optional[int] = tokenizer.eos_token_id
snake_case: Dict = 'speech_to_text_2'
snake_case: Optional[Any] = 'wav2vec2'
snake_case: Tuple = SpeechEncoderDecoderConfig.from_dict(__A )
hf_wavavec.save_pretrained(__A )
feature_extractor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
parser.add_argument("--dict_path", default=None, type=str, help="Path to dict of fine-tuned model")
parser.add_argument(
"--encoder_config_path",
default="facebook/wav2vec2-large-lv60",
type=str,
help="Path to hf encoder wav2vec2 checkpoint config",
)
parser.add_argument(
"--decoder_config_path",
default="facebook/s2t-small-mustc-en-fr-st",
type=str,
help="Path to hf decoder s2t checkpoint config",
)
parser.add_argument("--vocab_size", default=10_224, type=int, help="Vocab size of decoder")
parser.add_argument("--num_decoder_layers", default=7, type=int, help="Number of decoder layers")
__UpperCAmelCase = parser.parse_args()
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.dict_path,
encoder_config_path=args.encoder_config_path,
decoder_config_path=args.decoder_config_path,
vocab_size=args.vocab_size,
num_decoder_layers=args.num_decoder_layers,
) | 692 | 1 |
'''simple docstring'''
from math import factorial
__UpperCAmelCase = {str(d): factorial(d) for d in range(10)}
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
return sum(DIGIT_FACTORIAL[d] for d in str(__A ) )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = 7 * factorial(9 ) + 1
return sum(i for i in range(3 , __A ) if sum_of_digit_factorial(__A ) == i )
if __name__ == "__main__":
print(F'{solution() = }') | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 1_00 ):
'''simple docstring'''
snake_case: List[str] = n * (n + 1) * (2 * n + 1) / 6
snake_case: List[Any] = (n * (n + 1) / 2) ** 2
return int(square_of_sum - sum_of_squares )
if __name__ == "__main__":
print(F'{solution() = }') | 692 | 1 |
'''simple docstring'''
import argparse
import os
import torch
from transformers import (
XLNetConfig,
XLNetForQuestionAnswering,
XLNetForSequenceClassification,
XLNetLMHeadModel,
load_tf_weights_in_xlnet,
)
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
__UpperCAmelCase = {
"cola": 2,
"mnli": 3,
"mrpc": 2,
"sst-2": 2,
"sts-b": 1,
"qqp": 2,
"qnli": 2,
"rte": 2,
"wnli": 2,
}
logging.set_verbosity_info()
def lowerCAmelCase_ ( __A : int , __A : Tuple , __A : int , __A : List[Any]=None ):
'''simple docstring'''
snake_case: List[str] = XLNetConfig.from_json_file(__A )
snake_case: List[str] = finetuning_task.lower() if finetuning_task is not None else ''
if finetuning_task in GLUE_TASKS_NUM_LABELS:
print(f"""Building PyTorch XLNetForSequenceClassification model from configuration: {config}""" )
snake_case: int = finetuning_task
snake_case: List[Any] = GLUE_TASKS_NUM_LABELS[finetuning_task]
snake_case: Optional[int] = XLNetForSequenceClassification(__A )
elif "squad" in finetuning_task:
snake_case: Dict = finetuning_task
snake_case: Optional[Any] = XLNetForQuestionAnswering(__A )
else:
snake_case: Dict = XLNetLMHeadModel(__A )
# Load weights from tf checkpoint
load_tf_weights_in_xlnet(__A , __A , __A )
# Save pytorch-model
snake_case: Optional[Any] = os.path.join(__A , __A )
snake_case: List[Any] = os.path.join(__A , __A )
print(f"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(f"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , 'w' , encoding='utf-8' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--tf_checkpoint_path", default=None, type=str, required=True, help="Path to the TensorFlow checkpoint path."
)
parser.add_argument(
"--xlnet_config_file",
default=None,
type=str,
required=True,
help=(
"The config json file corresponding to the pre-trained XLNet model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--finetuning_task",
default=None,
type=str,
help="Name of a task on which the XLNet TensorFlow model was fine-tuned",
)
__UpperCAmelCase = parser.parse_args()
print(args)
convert_xlnet_checkpoint_to_pytorch(
args.tf_checkpoint_path, args.xlnet_config_file, args.pytorch_dump_folder_path, args.finetuning_task
) | 692 |
'''simple docstring'''
import argparse
from typing import Dict
import tensorflow as tf
import torch
from tqdm import tqdm
from transformers import BigBirdPegasusConfig, BigBirdPegasusForConditionalGeneration
__UpperCAmelCase = [
# tf -> hf
("/", "."),
("layer_", "layers."),
("kernel", "weight"),
("beta", "bias"),
("gamma", "weight"),
("pegasus", "model"),
]
__UpperCAmelCase = [
(".output.dense", ".fc2"),
("intermediate.LayerNorm", "final_layer_norm"),
("intermediate.dense", "fc1"),
]
__UpperCAmelCase = (
INIT_COMMON
+ [
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.out_proj"),
("attention.self", "self_attn"),
("attention.encdec.LayerNorm", "encoder_attn_layer_norm"),
("attention.encdec_output.dense", "encoder_attn.out_proj"),
("attention.encdec", "encoder_attn"),
("key", "k_proj"),
("value", "v_proj"),
("query", "q_proj"),
("decoder.LayerNorm", "decoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = (
INIT_COMMON
+ [
("embeddings.word_embeddings", "shared.weight"),
("embeddings.position_embeddings", "embed_positions.weight"),
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.output"),
("attention.self", "self_attn.self"),
("encoder.LayerNorm", "encoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = [
"encdec/key/bias",
"encdec/query/bias",
"encdec/value/bias",
"self/key/bias",
"self/query/bias",
"self/value/bias",
"encdec_output/dense/bias",
"attention/output/dense/bias",
]
def lowerCAmelCase_ ( __A : Dict , __A : List[Any] ):
'''simple docstring'''
for tf_name, hf_name in patterns:
snake_case: List[Any] = k.replace(__A , __A )
return k
def lowerCAmelCase_ ( __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[int] = BigBirdPegasusConfig(**__A )
snake_case: List[Any] = BigBirdPegasusForConditionalGeneration(__A )
snake_case: Any = torch_model.state_dict()
snake_case: Any = {}
# separating decoder weights
snake_case: Optional[Any] = {k: tf_weights[k] for k in tf_weights if k.startswith('pegasus/decoder' )}
snake_case: Any = {k: tf_weights[k] for k in tf_weights if not k.startswith('pegasus/decoder' )}
for k, v in tqdm(decoder_weights.items() , 'tf -> hf conversion' ):
snake_case: List[str] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Any = DECODER_PATTERNS
snake_case: int = rename_state_dict_key(__A , __A )
if new_k not in state_dict:
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: Optional[Any] = v.T
snake_case: Any = torch.from_numpy(__A )
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
for k, v in tqdm(remaining_weights.items() , 'tf -> hf conversion' ):
snake_case: List[Any] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Union[str, Any] = REMAINING_PATTERNS
snake_case: str = rename_state_dict_key(__A , __A )
if new_k not in state_dict and k != "pegasus/embeddings/position_embeddings":
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: int = v.T
snake_case: Any = torch.from_numpy(__A )
if k != "pegasus/embeddings/position_embeddings":
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
snake_case: str = mapping['model.embed_positions.weight']
snake_case: Any = mapping.pop('model.embed_positions.weight' )
snake_case , snake_case: Union[str, Any] = torch_model.load_state_dict(__A , strict=__A )
snake_case: Optional[int] = [
k
for k in missing
if k
not in [
'final_logits_bias',
'model.encoder.embed_tokens.weight',
'model.decoder.embed_tokens.weight',
'lm_head.weight',
]
]
assert unexpected_missing == [], f"""no matches found for the following torch keys {unexpected_missing}"""
assert extra == [], f"""no matches found for the following tf keys {extra}"""
return torch_model
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
snake_case: Tuple = tf.train.list_variables(__A )
snake_case: str = {}
snake_case: List[str] = ['global_step']
for name, shape in tqdm(__A , desc='converting tf checkpoint to dict' ):
snake_case: str = any(pat in name for pat in ignore_name )
if skip_key:
continue
snake_case: Any = tf.train.load_variable(__A , __A )
snake_case: Optional[int] = array
return tf_weights
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict ):
'''simple docstring'''
snake_case: int = get_tf_weights_as_numpy(__A )
snake_case: int = convert_bigbird_pegasus(__A , __A )
torch_model.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--tf_ckpt_path", type=str, help="passed to tf.train.list_variables")
parser.add_argument("--save_dir", default=None, type=str, help="Path to the output PyTorch model.")
__UpperCAmelCase = parser.parse_args()
__UpperCAmelCase = {}
convert_bigbird_pegasus_ckpt_to_pytorch(args.tf_ckpt_path, args.save_dir, config_update=config_update) | 692 | 1 |
'''simple docstring'''
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModelWithProjection, CLIPTokenizer
from diffusers import HeunDiscreteScheduler, PriorTransformer, ShapEPipeline
from diffusers.pipelines.shap_e import ShapERenderer
from diffusers.utils import load_numpy, slow
from diffusers.utils.testing_utils import require_torch_gpu, torch_device
from ..test_pipelines_common import PipelineTesterMixin, assert_mean_pixel_difference
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ShapEPipeline
__UpperCamelCase = ["prompt"]
__UpperCamelCase = ["prompt"]
__UpperCamelCase = [
"num_images_per_prompt",
"num_inference_steps",
"generator",
"latents",
"guidance_scale",
"frame_size",
"output_type",
"return_dict",
]
__UpperCamelCase = False
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 32
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 32
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.time_input_dim * 4
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 8
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
return tokenizer
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Dict = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=self.text_embedder_hidden_size , projection_dim=self.text_embedder_hidden_size , intermediate_size=37 , layer_norm_eps=1E-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=10_00 , )
return CLIPTextModelWithProjection(SCREAMING_SNAKE_CASE__ )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Dict = {
'num_attention_heads': 2,
'attention_head_dim': 16,
'embedding_dim': self.time_input_dim,
'num_embeddings': 32,
'embedding_proj_dim': self.text_embedder_hidden_size,
'time_embed_dim': self.time_embed_dim,
'num_layers': 1,
'clip_embed_dim': self.time_input_dim * 2,
'additional_embeddings': 0,
'time_embed_act_fn': 'gelu',
'norm_in_type': 'layer',
'encoder_hid_proj_type': None,
'added_emb_type': None,
}
snake_case: Optional[Any] = PriorTransformer(**SCREAMING_SNAKE_CASE__ )
return model
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
torch.manual_seed(0 )
snake_case: Tuple = {
'param_shapes': (
(self.renderer_dim, 93),
(self.renderer_dim, 8),
(self.renderer_dim, 8),
(self.renderer_dim, 8),
),
'd_latent': self.time_input_dim,
'd_hidden': self.renderer_dim,
'n_output': 12,
'background': (
0.1,
0.1,
0.1,
),
}
snake_case: Union[str, Any] = ShapERenderer(**SCREAMING_SNAKE_CASE__ )
return model
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.dummy_prior
snake_case: Dict = self.dummy_text_encoder
snake_case: int = self.dummy_tokenizer
snake_case: List[str] = self.dummy_renderer
snake_case: Tuple = HeunDiscreteScheduler(
beta_schedule='exp' , num_train_timesteps=10_24 , prediction_type='sample' , use_karras_sigmas=SCREAMING_SNAKE_CASE__ , clip_sample=SCREAMING_SNAKE_CASE__ , clip_sample_range=1.0 , )
snake_case: Any = {
'prior': prior,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'renderer': renderer,
'scheduler': scheduler,
}
return components
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0 ):
'''simple docstring'''
if str(SCREAMING_SNAKE_CASE__ ).startswith('mps' ):
snake_case: Union[str, Any] = torch.manual_seed(SCREAMING_SNAKE_CASE__ )
else:
snake_case: Union[str, Any] = torch.Generator(device=SCREAMING_SNAKE_CASE__ ).manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: int = {
'prompt': 'horse',
'generator': generator,
'num_inference_steps': 1,
'frame_size': 32,
'output_type': 'np',
}
return inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = 'cpu'
snake_case: List[str] = self.get_dummy_components()
snake_case: Tuple = self.pipeline_class(**SCREAMING_SNAKE_CASE__ )
snake_case: int = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = pipe(**self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = output.images[0]
snake_case: List[str] = image[0, -3:, -3:, -1]
assert image.shape == (20, 32, 32, 3)
snake_case: Union[str, Any] = np.array(
[
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
0.00_03_92_16,
] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
self._test_inference_batch_consistent(batch_sizes=[1, 2] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = torch_device == 'cpu'
snake_case: Union[str, Any] = True
self._test_inference_batch_single_identical(
batch_size=2 , test_max_difference=SCREAMING_SNAKE_CASE__ , relax_max_difference=SCREAMING_SNAKE_CASE__ , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_dummy_components()
snake_case: Dict = self.pipeline_class(**SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = 1
snake_case: List[Any] = 2
snake_case: Optional[Any] = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
for key in inputs.keys():
if key in self.batch_params:
snake_case: Any = batch_size * [inputs[key]]
snake_case: Dict = pipe(**SCREAMING_SNAKE_CASE__ , num_images_per_prompt=SCREAMING_SNAKE_CASE__ )[0]
assert images.shape[0] == batch_size * num_images_per_prompt
@slow
@require_torch_gpu
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'
'/shap_e/test_shap_e_np_out.npy' )
snake_case: Tuple = ShapEPipeline.from_pretrained('openai/shap-e' )
snake_case: List[str] = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = torch.Generator(device=SCREAMING_SNAKE_CASE__ ).manual_seed(0 )
snake_case: int = pipe(
'a shark' , generator=SCREAMING_SNAKE_CASE__ , guidance_scale=15.0 , num_inference_steps=64 , frame_size=64 , output_type='np' , ).images[0]
assert images.shape == (20, 64, 64, 3)
assert_mean_pixel_difference(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: str = [0] * len(__A )
snake_case: Tuple = []
snake_case: Tuple = [1] * len(__A )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(__A ) ):
if indegree[i] == 0:
queue.append(__A )
while queue:
snake_case: int = queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
snake_case: Any = long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(__A )
print(max(__A ) )
# Adjacency list of Graph
__UpperCAmelCase = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph) | 692 | 1 |
'''simple docstring'''
from collections import OrderedDict
from ...utils import logging
from .auto_factory import _BaseAutoModelClass, _LazyAutoMapping, auto_class_update
from .configuration_auto import CONFIG_MAPPING_NAMES
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = OrderedDict(
[
# Base model mapping
("albert", "FlaxAlbertModel"),
("bart", "FlaxBartModel"),
("beit", "FlaxBeitModel"),
("bert", "FlaxBertModel"),
("big_bird", "FlaxBigBirdModel"),
("blenderbot", "FlaxBlenderbotModel"),
("blenderbot-small", "FlaxBlenderbotSmallModel"),
("clip", "FlaxCLIPModel"),
("distilbert", "FlaxDistilBertModel"),
("electra", "FlaxElectraModel"),
("gpt-sw3", "FlaxGPT2Model"),
("gpt2", "FlaxGPT2Model"),
("gpt_neo", "FlaxGPTNeoModel"),
("gptj", "FlaxGPTJModel"),
("longt5", "FlaxLongT5Model"),
("marian", "FlaxMarianModel"),
("mbart", "FlaxMBartModel"),
("mt5", "FlaxMT5Model"),
("opt", "FlaxOPTModel"),
("pegasus", "FlaxPegasusModel"),
("regnet", "FlaxRegNetModel"),
("resnet", "FlaxResNetModel"),
("roberta", "FlaxRobertaModel"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormModel"),
("roformer", "FlaxRoFormerModel"),
("t5", "FlaxT5Model"),
("vision-text-dual-encoder", "FlaxVisionTextDualEncoderModel"),
("vit", "FlaxViTModel"),
("wav2vec2", "FlaxWav2Vec2Model"),
("whisper", "FlaxWhisperModel"),
("xglm", "FlaxXGLMModel"),
("xlm-roberta", "FlaxXLMRobertaModel"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for pre-training mapping
("albert", "FlaxAlbertForPreTraining"),
("bart", "FlaxBartForConditionalGeneration"),
("bert", "FlaxBertForPreTraining"),
("big_bird", "FlaxBigBirdForPreTraining"),
("electra", "FlaxElectraForPreTraining"),
("longt5", "FlaxLongT5ForConditionalGeneration"),
("mbart", "FlaxMBartForConditionalGeneration"),
("mt5", "FlaxMT5ForConditionalGeneration"),
("roberta", "FlaxRobertaForMaskedLM"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMaskedLM"),
("roformer", "FlaxRoFormerForMaskedLM"),
("t5", "FlaxT5ForConditionalGeneration"),
("wav2vec2", "FlaxWav2Vec2ForPreTraining"),
("whisper", "FlaxWhisperForConditionalGeneration"),
("xlm-roberta", "FlaxXLMRobertaForMaskedLM"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Masked LM mapping
("albert", "FlaxAlbertForMaskedLM"),
("bart", "FlaxBartForConditionalGeneration"),
("bert", "FlaxBertForMaskedLM"),
("big_bird", "FlaxBigBirdForMaskedLM"),
("distilbert", "FlaxDistilBertForMaskedLM"),
("electra", "FlaxElectraForMaskedLM"),
("mbart", "FlaxMBartForConditionalGeneration"),
("roberta", "FlaxRobertaForMaskedLM"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMaskedLM"),
("roformer", "FlaxRoFormerForMaskedLM"),
("xlm-roberta", "FlaxXLMRobertaForMaskedLM"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Seq2Seq Causal LM mapping
("bart", "FlaxBartForConditionalGeneration"),
("blenderbot", "FlaxBlenderbotForConditionalGeneration"),
("blenderbot-small", "FlaxBlenderbotSmallForConditionalGeneration"),
("encoder-decoder", "FlaxEncoderDecoderModel"),
("longt5", "FlaxLongT5ForConditionalGeneration"),
("marian", "FlaxMarianMTModel"),
("mbart", "FlaxMBartForConditionalGeneration"),
("mt5", "FlaxMT5ForConditionalGeneration"),
("pegasus", "FlaxPegasusForConditionalGeneration"),
("t5", "FlaxT5ForConditionalGeneration"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Image-classsification
("beit", "FlaxBeitForImageClassification"),
("regnet", "FlaxRegNetForImageClassification"),
("resnet", "FlaxResNetForImageClassification"),
("vit", "FlaxViTForImageClassification"),
]
)
__UpperCAmelCase = OrderedDict(
[
("vision-encoder-decoder", "FlaxVisionEncoderDecoderModel"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Causal LM mapping
("bart", "FlaxBartForCausalLM"),
("bert", "FlaxBertForCausalLM"),
("big_bird", "FlaxBigBirdForCausalLM"),
("electra", "FlaxElectraForCausalLM"),
("gpt-sw3", "FlaxGPT2LMHeadModel"),
("gpt2", "FlaxGPT2LMHeadModel"),
("gpt_neo", "FlaxGPTNeoForCausalLM"),
("gptj", "FlaxGPTJForCausalLM"),
("opt", "FlaxOPTForCausalLM"),
("roberta", "FlaxRobertaForCausalLM"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForCausalLM"),
("xglm", "FlaxXGLMForCausalLM"),
("xlm-roberta", "FlaxXLMRobertaForCausalLM"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Sequence Classification mapping
("albert", "FlaxAlbertForSequenceClassification"),
("bart", "FlaxBartForSequenceClassification"),
("bert", "FlaxBertForSequenceClassification"),
("big_bird", "FlaxBigBirdForSequenceClassification"),
("distilbert", "FlaxDistilBertForSequenceClassification"),
("electra", "FlaxElectraForSequenceClassification"),
("mbart", "FlaxMBartForSequenceClassification"),
("roberta", "FlaxRobertaForSequenceClassification"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForSequenceClassification"),
("roformer", "FlaxRoFormerForSequenceClassification"),
("xlm-roberta", "FlaxXLMRobertaForSequenceClassification"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Question Answering mapping
("albert", "FlaxAlbertForQuestionAnswering"),
("bart", "FlaxBartForQuestionAnswering"),
("bert", "FlaxBertForQuestionAnswering"),
("big_bird", "FlaxBigBirdForQuestionAnswering"),
("distilbert", "FlaxDistilBertForQuestionAnswering"),
("electra", "FlaxElectraForQuestionAnswering"),
("mbart", "FlaxMBartForQuestionAnswering"),
("roberta", "FlaxRobertaForQuestionAnswering"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForQuestionAnswering"),
("roformer", "FlaxRoFormerForQuestionAnswering"),
("xlm-roberta", "FlaxXLMRobertaForQuestionAnswering"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Token Classification mapping
("albert", "FlaxAlbertForTokenClassification"),
("bert", "FlaxBertForTokenClassification"),
("big_bird", "FlaxBigBirdForTokenClassification"),
("distilbert", "FlaxDistilBertForTokenClassification"),
("electra", "FlaxElectraForTokenClassification"),
("roberta", "FlaxRobertaForTokenClassification"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForTokenClassification"),
("roformer", "FlaxRoFormerForTokenClassification"),
("xlm-roberta", "FlaxXLMRobertaForTokenClassification"),
]
)
__UpperCAmelCase = OrderedDict(
[
# Model for Multiple Choice mapping
("albert", "FlaxAlbertForMultipleChoice"),
("bert", "FlaxBertForMultipleChoice"),
("big_bird", "FlaxBigBirdForMultipleChoice"),
("distilbert", "FlaxDistilBertForMultipleChoice"),
("electra", "FlaxElectraForMultipleChoice"),
("roberta", "FlaxRobertaForMultipleChoice"),
("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMultipleChoice"),
("roformer", "FlaxRoFormerForMultipleChoice"),
("xlm-roberta", "FlaxXLMRobertaForMultipleChoice"),
]
)
__UpperCAmelCase = OrderedDict(
[
("bert", "FlaxBertForNextSentencePrediction"),
]
)
__UpperCAmelCase = OrderedDict(
[
("speech-encoder-decoder", "FlaxSpeechEncoderDecoderModel"),
("whisper", "FlaxWhisperForConditionalGeneration"),
]
)
__UpperCAmelCase = OrderedDict(
[
("whisper", "FlaxWhisperForAudioClassification"),
]
)
__UpperCAmelCase = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_MAPPING_NAMES)
__UpperCAmelCase = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_PRETRAINING_MAPPING_NAMES)
__UpperCAmelCase = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_MASKED_LM_MAPPING_NAMES)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_VISION_2_SEQ_MAPPING_NAMES)
__UpperCAmelCase = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_CAUSAL_LM_MAPPING_NAMES)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_QUESTION_ANSWERING_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_MULTIPLE_CHOICE_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES
)
__UpperCAmelCase = _LazyAutoMapping(
CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModel)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_PRETRAINING_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForPreTraining, head_doc="pretraining")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_CAUSAL_LM_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForCausalLM, head_doc="causal language modeling")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_MASKED_LM_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForMaskedLM, head_doc="masked language modeling")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForSeqaSeqLM, head_doc="sequence-to-sequence language modeling", checkpoint_for_example="t5-base"
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForSequenceClassification, head_doc="sequence classification"
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_QUESTION_ANSWERING_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForQuestionAnswering, head_doc="question answering")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForTokenClassification, head_doc="token classification"
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_MULTIPLE_CHOICE_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForMultipleChoice, head_doc="multiple choice")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForNextSentencePrediction, head_doc="next sentence prediction"
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForImageClassification, head_doc="image classification"
)
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_VISION_2_SEQ_MAPPING
__UpperCAmelCase = auto_class_update(FlaxAutoModelForVisionaSeq, head_doc="vision-to-text modeling")
class SCREAMING_SNAKE_CASE ( _BaseAutoModelClass ):
'''simple docstring'''
__UpperCamelCase = FLAX_MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING
__UpperCAmelCase = auto_class_update(
FlaxAutoModelForSpeechSeqaSeq, head_doc="sequence-to-sequence speech-to-text modeling"
) | 692 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
import numpy as np
import pytest
from transformers import BertTokenizer, BertTokenizerFast
from transformers.models.bert.tokenization_bert import VOCAB_FILES_NAMES
from transformers.testing_utils import require_vision
from transformers.utils import FEATURE_EXTRACTOR_NAME, is_vision_available
if is_vision_available():
from PIL import Image
from transformers import ChineseCLIPImageProcessor, ChineseCLIPProcessor
@require_vision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = tempfile.mkdtemp()
snake_case: Optional[Any] = [
'[UNK]',
'[CLS]',
'[SEP]',
'[PAD]',
'[MASK]',
'的',
'价',
'格',
'是',
'15',
'便',
'alex',
'##andra',
'，',
'。',
'-',
't',
'shirt',
]
snake_case: Optional[Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
snake_case: Optional[int] = {
'do_resize': True,
'size': {'height': 2_24, 'width': 2_24},
'do_center_crop': True,
'crop_size': {'height': 18, 'width': 18},
'do_normalize': True,
'image_mean': [0.48_14_54_66, 0.4_57_82_75, 0.40_82_10_73],
'image_std': [0.26_86_29_54, 0.26_13_02_58, 0.27_57_77_11],
'do_convert_rgb': True,
}
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , SCREAMING_SNAKE_CASE__ )
with open(self.image_processor_file , 'w' , encoding='utf-8' ) as fp:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizer.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizerFast.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return ChineseCLIPImageProcessor.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Tuple = [Image.fromarray(np.moveaxis(SCREAMING_SNAKE_CASE__ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizer()
snake_case: Union[str, Any] = self.get_rust_tokenizer()
snake_case: Union[str, Any] = self.get_image_processor()
snake_case: List[str] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_slow.save_pretrained(self.tmpdirname )
snake_case: List[str] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname , use_fast=SCREAMING_SNAKE_CASE__ )
snake_case: Any = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_fast.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor_slow.tokenizer.get_vocab() , tokenizer_slow.get_vocab() )
self.assertEqual(processor_fast.tokenizer.get_vocab() , tokenizer_fast.get_vocab() )
self.assertEqual(tokenizer_slow.get_vocab() , tokenizer_fast.get_vocab() )
self.assertIsInstance(processor_slow.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor_slow.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertEqual(processor_fast.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertIsInstance(processor_slow.image_processor , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=self.get_tokenizer() , image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = self.get_tokenizer(cls_token='(CLS)' , sep_token='(SEP)' )
snake_case: Union[str, Any] = self.get_image_processor(do_normalize=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = ChineseCLIPProcessor.from_pretrained(
self.tmpdirname , cls_token='(CLS)' , sep_token='(SEP)' , do_normalize=SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
self.assertIsInstance(processor.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_image_processor()
snake_case: Tuple = self.get_tokenizer()
snake_case: Optional[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.prepare_image_inputs()
snake_case: List[Any] = image_processor(SCREAMING_SNAKE_CASE__ , return_tensors='np' )
snake_case: Dict = processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='np' )
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_image_processor()
snake_case: Optional[int] = self.get_tokenizer()
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Union[str, Any] = processor(text=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer(SCREAMING_SNAKE_CASE__ )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Tuple = self.prepare_image_inputs()
snake_case: Any = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , ['input_ids', 'token_type_ids', 'attention_mask', 'pixel_values'] )
# test if it raises when no input is passed
with pytest.raises(SCREAMING_SNAKE_CASE__ ):
processor()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.get_image_processor()
snake_case: str = self.get_tokenizer()
snake_case: Union[str, Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = [[1, 4, 5, 8, 1, 0, 8], [3, 4, 3, 1, 1, 8, 9]]
snake_case: int = processor.batch_decode(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = 'Alexandra，T-shirt的价格是15便士。'
snake_case: List[Any] = self.prepare_image_inputs()
snake_case: Dict = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , processor.model_input_names ) | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
__UpperCAmelCase = "#"
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self ):
'''simple docstring'''
snake_case: dict = {}
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: str = self._trie
for char in text:
if char not in trie:
snake_case: Optional[int] = {}
snake_case: Any = trie[char]
snake_case: Optional[Any] = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: str = self._trie
for char in prefix:
if char in trie:
snake_case: int = trie[char]
else:
return []
return self._elements(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = []
for c, v in d.items():
snake_case: Union[str, Any] = [' '] if c == END else [(c + s) for s in self._elements(SCREAMING_SNAKE_CASE__ )]
result.extend(SCREAMING_SNAKE_CASE__ )
return tuple(SCREAMING_SNAKE_CASE__ )
__UpperCAmelCase = Trie()
__UpperCAmelCase = ("depart", "detergent", "daring", "dog", "deer", "deal")
for word in words:
trie.insert_word(word)
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[str] = trie.find_word(__A )
return tuple(string + word for word in suffixes )
def lowerCAmelCase_ ( ):
'''simple docstring'''
print(autocomplete_using_trie('de' ) )
if __name__ == "__main__":
import doctest
doctest.testmod()
main() | 692 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/swinv2-tiny-patch4-window8-256": (
"https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256/resolve/main/config.json"
),
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "swinv2"
__UpperCamelCase = {
"num_attention_heads": "num_heads",
"num_hidden_layers": "num_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=96 , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[3, 6, 12, 24] , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=32 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: int = image_size
snake_case: Union[str, Any] = patch_size
snake_case: List[str] = num_channels
snake_case: Tuple = embed_dim
snake_case: str = depths
snake_case: Any = len(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = num_heads
snake_case: Optional[int] = window_size
snake_case: Any = mlp_ratio
snake_case: Optional[int] = qkv_bias
snake_case: Union[str, Any] = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: Dict = drop_path_rate
snake_case: List[str] = hidden_act
snake_case: int = use_absolute_embeddings
snake_case: Any = layer_norm_eps
snake_case: Dict = initializer_range
snake_case: List[Any] = encoder_stride
# we set the hidden_size attribute in order to make Swinv2 work with VisionEncoderDecoderModel
# this indicates the channel dimension after the last stage of the model
snake_case: Tuple = int(embed_dim * 2 ** (len(SCREAMING_SNAKE_CASE__ ) - 1) )
snake_case: Union[str, Any] = (0, 0, 0, 0) | 692 | 1 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import is_flaky, require_torch, require_vision
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
from transformers import DonutImageProcessor
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=18 , SCREAMING_SNAKE_CASE__=30 , SCREAMING_SNAKE_CASE__=4_00 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=[0.5, 0.5, 0.5] , SCREAMING_SNAKE_CASE__=[0.5, 0.5, 0.5] , ):
'''simple docstring'''
snake_case: Optional[int] = parent
snake_case: Union[str, Any] = batch_size
snake_case: List[str] = num_channels
snake_case: str = image_size
snake_case: Union[str, Any] = min_resolution
snake_case: Dict = max_resolution
snake_case: List[Any] = do_resize
snake_case: Dict = size if size is not None else {'height': 18, 'width': 20}
snake_case: List[str] = do_thumbnail
snake_case: Optional[int] = do_align_axis
snake_case: Union[str, Any] = do_pad
snake_case: Dict = do_normalize
snake_case: Tuple = image_mean
snake_case: str = image_std
def _UpperCamelCase ( self ):
'''simple docstring'''
return {
"do_resize": self.do_resize,
"size": self.size,
"do_thumbnail": self.do_thumbnail,
"do_align_long_axis": self.do_align_axis,
"do_pad": self.do_pad,
"do_normalize": self.do_normalize,
"image_mean": self.image_mean,
"image_std": self.image_std,
}
@require_torch
@require_vision
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = DonutImageProcessor if is_vision_available() else None
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = DonutImageProcessingTester(self )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.image_processor_tester.prepare_image_processor_dict()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'do_resize' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'size' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'do_thumbnail' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'do_align_long_axis' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'do_pad' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'do_normalize' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'image_mean' ) )
self.assertTrue(hasattr(SCREAMING_SNAKE_CASE__ , 'image_std' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.image_processing_class.from_dict(self.image_processor_dict )
self.assertEqual(image_processor.size , {'height': 18, 'width': 20} )
snake_case: Any = self.image_processing_class.from_dict(self.image_processor_dict , size=42 )
self.assertEqual(image_processor.size , {'height': 42, 'width': 42} )
# Previous config had dimensions in (width, height) order
snake_case: Optional[Any] = self.image_processing_class.from_dict(self.image_processor_dict , size=(42, 84) )
self.assertEqual(image_processor.size , {'height': 84, 'width': 42} )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@is_flaky()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
snake_case: Tuple = prepare_image_inputs(self.image_processor_tester , equal_resolution=SCREAMING_SNAKE_CASE__ )
for image in image_inputs:
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , Image.Image )
# Test not batched input
snake_case: str = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , )
# Test batched
snake_case: str = image_processing(SCREAMING_SNAKE_CASE__ , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , )
@is_flaky()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.image_processing_class(**self.image_processor_dict )
# create random numpy tensors
snake_case: List[Any] = prepare_image_inputs(self.image_processor_tester , equal_resolution=SCREAMING_SNAKE_CASE__ , numpify=SCREAMING_SNAKE_CASE__ )
for image in image_inputs:
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , np.ndarray )
# Test not batched input
snake_case: Dict = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , )
# Test batched
snake_case: Optional[Any] = image_processing(SCREAMING_SNAKE_CASE__ , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , )
@is_flaky()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.image_processing_class(**self.image_processor_dict )
# create random PyTorch tensors
snake_case: str = prepare_image_inputs(self.image_processor_tester , equal_resolution=SCREAMING_SNAKE_CASE__ , torchify=SCREAMING_SNAKE_CASE__ )
for image in image_inputs:
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , torch.Tensor )
# Test not batched input
snake_case: List[Any] = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , )
# Test batched
snake_case: Optional[int] = image_processing(SCREAMING_SNAKE_CASE__ , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.size['height'],
self.image_processor_tester.size['width'],
) , ) | 692 |
'''simple docstring'''
import os
import sys
import unittest
__UpperCAmelCase = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__))))
sys.path.append(os.path.join(git_repo_path, "utils"))
import check_dummies # noqa: E402
from check_dummies import create_dummy_files, create_dummy_object, find_backend, read_init # noqa: E402
# Align TRANSFORMERS_PATH in check_dummies with the current path
__UpperCAmelCase = os.path.join(git_repo_path, "src", "transformers")
__UpperCAmelCase = "\n{0} = None\n"
__UpperCAmelCase = "\nclass {0}(metaclass=DummyObject):\n _backends = {1}\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, {1})\n"
__UpperCAmelCase = "\ndef {0}(*args, **kwargs):\n requires_backends({0}, {1})\n"
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = find_backend(' _import_structure["models.albert"].append("AlbertTokenizerFast")' )
self.assertIsNone(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = find_backend(' if not is_tokenizers_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tokenizers' )
snake_case: List[Any] = find_backend(' if not is_tensorflow_text_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tensorflow_text' )
snake_case: int = find_backend(' if not (is_sentencepiece_available() and is_tokenizers_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers' )
snake_case: Optional[Any] = find_backend(
' if not (is_sentencepiece_available() and is_tensorflow_text_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tensorflow_text' )
snake_case: Dict = find_backend(
' if not (is_sentencepiece_available() and is_tokenizers_available() and is_vision_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers_and_vision' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = read_init()
# We don't assert on the exact list of keys to allow for smooth grow of backend-specific objects
self.assertIn('torch' , SCREAMING_SNAKE_CASE__ )
self.assertIn('tensorflow_text' , SCREAMING_SNAKE_CASE__ )
self.assertIn('sentencepiece_and_tokenizers' , SCREAMING_SNAKE_CASE__ )
# Likewise, we can't assert on the exact content of a key
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertModel' , objects['tf'] )
self.assertIn('FlaxBertModel' , objects['flax'] )
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertTokenizer' , objects['tensorflow_text'] )
self.assertIn('convert_slow_tokenizer' , objects['sentencepiece_and_tokenizers'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = create_dummy_object('CONSTANT' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , '\nCONSTANT = None\n' )
snake_case: Any = create_dummy_object('function' , '\'torch\'' )
self.assertEqual(
SCREAMING_SNAKE_CASE__ , '\ndef function(*args, **kwargs):\n requires_backends(function, \'torch\')\n' )
snake_case: Optional[int] = '\nclass FakeClass(metaclass=DummyObject):\n _backends = \'torch\'\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, \'torch\')\n'
snake_case: Tuple = create_dummy_object('FakeClass' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = '# This file is autogenerated by the command `make fix-copies`, do not edit.\nfrom ..utils import DummyObject, requires_backends\n\n\nCONSTANT = None\n\n\ndef function(*args, **kwargs):\n requires_backends(function, ["torch"])\n\n\nclass FakeClass(metaclass=DummyObject):\n _backends = ["torch"]\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, ["torch"])\n'
snake_case: Optional[int] = create_dummy_files({'torch': ['CONSTANT', 'function', 'FakeClass']} )
self.assertEqual(dummy_files['torch'] , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/trocr-base-handwritten": (
"https://huggingface.co/microsoft/trocr-base-handwritten/resolve/main/config.json"
),
# See all TrOCR models at https://huggingface.co/models?filter=trocr
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "trocr"
__UpperCamelCase = ["past_key_values"]
__UpperCamelCase = {
"num_attention_heads": "decoder_attention_heads",
"hidden_size": "d_model",
"num_hidden_layers": "decoder_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=5_02_65 , SCREAMING_SNAKE_CASE__=10_24 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=40_96 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=1 , SCREAMING_SNAKE_CASE__=0 , SCREAMING_SNAKE_CASE__=2 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: str = vocab_size
snake_case: Dict = d_model
snake_case: Optional[int] = decoder_layers
snake_case: Union[str, Any] = decoder_attention_heads
snake_case: int = decoder_ffn_dim
snake_case: Tuple = activation_function
snake_case: int = max_position_embeddings
snake_case: Any = dropout
snake_case: List[str] = attention_dropout
snake_case: Optional[int] = activation_dropout
snake_case: Dict = init_std
snake_case: Optional[Any] = decoder_layerdrop
snake_case: Tuple = use_cache
snake_case: Tuple = scale_embedding
snake_case: List[Any] = use_learned_position_embeddings
snake_case: List[Any] = layernorm_embedding
super().__init__(
pad_token_id=SCREAMING_SNAKE_CASE__ , bos_token_id=SCREAMING_SNAKE_CASE__ , eos_token_id=SCREAMING_SNAKE_CASE__ , decoder_start_token_id=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , ) | 692 |
'''simple docstring'''
import os
import warnings
from typing import List, Optional
from ...tokenization_utils_base import BatchEncoding
from ...utils import logging
from .configuration_rag import RagConfig
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = question_encoder
snake_case: Union[str, Any] = generator
snake_case: Optional[int] = self.question_encoder
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if os.path.isfile(SCREAMING_SNAKE_CASE__ ):
raise ValueError(F"""Provided path ({save_directory}) should be a directory, not a file""" )
os.makedirs(SCREAMING_SNAKE_CASE__ , exist_ok=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'question_encoder_tokenizer' )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'generator_tokenizer' )
self.question_encoder.save_pretrained(SCREAMING_SNAKE_CASE__ )
self.generator.save_pretrained(SCREAMING_SNAKE_CASE__ )
@classmethod
def _UpperCamelCase ( cls , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
from ..auto.tokenization_auto import AutoTokenizer
snake_case: int = kwargs.pop('config' , SCREAMING_SNAKE_CASE__ )
if config is None:
snake_case: str = RagConfig.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.question_encoder , subfolder='question_encoder_tokenizer' )
snake_case: Dict = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.generator , subfolder='generator_tokenizer' )
return cls(question_encoder=SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ )
def __call__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.current_tokenizer(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.batch_decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.question_encoder
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.generator
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "longest" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
warnings.warn(
'`prepare_seq2seq_batch` is deprecated and will be removed in version 5 of 🤗 Transformers. Use the '
'regular `__call__` method to prepare your inputs and the tokenizer under the `with_target_tokenizer` '
'context manager to prepare your targets. See the documentation of your specific tokenizer for more '
'details' , SCREAMING_SNAKE_CASE__ , )
if max_length is None:
snake_case: Optional[Any] = self.current_tokenizer.model_max_length
snake_case: int = self(
SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if tgt_texts is None:
return model_inputs
# Process tgt_texts
if max_target_length is None:
snake_case: Any = self.current_tokenizer.model_max_length
snake_case: List[str] = self(
text_target=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: Dict = labels['input_ids']
return model_inputs | 692 | 1 |
'''simple docstring'''
import argparse
import json
import os
import fairseq
import torch
from torch import nn
from transformers import (
SpeechaTextaConfig,
SpeechaTextaForCausalLM,
SpeechaTextaTokenizer,
SpeechEncoderDecoderConfig,
SpeechEncoderDecoderModel,
WavaVecaConfig,
WavaVecaFeatureExtractor,
WavaVecaModel,
logging,
)
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"post_extract_proj": "feature_projection.projection",
"encoder.pos_conv.0": "encoder.pos_conv_embed.conv",
"self_attn.k_proj": "encoder.layers.*.attention.k_proj",
"self_attn.v_proj": "encoder.layers.*.attention.v_proj",
"self_attn.q_proj": "encoder.layers.*.attention.q_proj",
"self_attn.out_proj": "encoder.layers.*.attention.out_proj",
"self_attn_layer_norm": "encoder.layers.*.layer_norm",
"fc1": "encoder.layers.*.feed_forward.intermediate_dense",
"fc2": "encoder.layers.*.feed_forward.output_dense",
"final_layer_norm": "encoder.layers.*.final_layer_norm",
"encoder.layer_norm": "encoder.layer_norm",
"w2v_model.layer_norm": "feature_projection.layer_norm",
"quantizer.weight_proj": "quantizer.weight_proj",
"quantizer.vars": "quantizer.codevectors",
"project_q": "project_q",
"final_proj": "project_hid",
"w2v_encoder.proj": "lm_head",
"mask_emb": "masked_spec_embed",
}
__UpperCAmelCase = [
"lm_head",
"quantizer.weight_proj",
"quantizer.codevectors",
"project_q",
"project_hid",
]
def lowerCAmelCase_ ( __A : Any , __A : Optional[Any] , __A : Union[str, Any] , __A : int , __A : Optional[int] ):
'''simple docstring'''
for attribute in key.split('.' ):
snake_case: List[str] = getattr(__A , __A )
if weight_type is not None:
snake_case: Optional[int] = getattr(__A , __A ).shape
else:
snake_case: Optional[int] = hf_pointer.shape
assert hf_shape == value.shape, (
f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
snake_case: Optional[int] = value
elif weight_type == "weight_g":
snake_case: List[str] = value
elif weight_type == "weight_v":
snake_case: Dict = value
elif weight_type == "bias":
snake_case: Optional[Any] = value
else:
snake_case: int = value
logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def lowerCAmelCase_ ( __A : List[Any] , __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = []
snake_case: List[Any] = fairseq_model.state_dict()
snake_case: Union[str, Any] = hf_model.feature_extractor
# if encoder has different dim to decoder -> use proj_weight
snake_case: Dict = None
for name, value in fairseq_dict.items():
snake_case: Tuple = False
if "conv_layers" in name:
load_conv_layer(
__A , __A , __A , __A , hf_model.config.feat_extract_norm == 'group' , )
snake_case: List[Any] = True
elif name.split('.' )[0] == "proj":
snake_case: List[Any] = fairseq_model.proj
snake_case: int = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
snake_case: int = True
if "*" in mapped_key:
snake_case: List[str] = name.split(__A )[0].split('.' )[-2]
snake_case: Dict = mapped_key.replace('*' , __A )
if "weight_g" in name:
snake_case: Tuple = 'weight_g'
elif "weight_v" in name:
snake_case: int = 'weight_v'
elif "bias" in name:
snake_case: Tuple = 'bias'
elif "weight" in name:
snake_case: List[Any] = 'weight'
else:
snake_case: Any = None
set_recursively(__A , __A , __A , __A , __A )
continue
if not is_used:
unused_weights.append(__A )
logger.warning(f"""Unused weights: {unused_weights}""" )
return proj_weight
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : int , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = full_name.split('conv_layers.' )[-1]
snake_case: Tuple = name.split('.' )
snake_case: Any = int(items[0] )
snake_case: Optional[int] = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
snake_case: Tuple = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
snake_case: int = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
snake_case: Any = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
snake_case: str = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(__A )
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case , snake_case: List[Any] = emb.weight.shape
snake_case: Optional[int] = nn.Linear(__A , __A , bias=__A )
snake_case: Any = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
with open(__A , 'r' , encoding='utf-8' ) as f:
snake_case: List[Any] = f.readlines()
snake_case: Any = [line.split(' ' )[0] for line in lines]
snake_case: int = len(__A )
snake_case: Dict = {
'<s>': 0,
'<pad>': 1,
'</s>': 2,
'<unk>': 3,
}
vocab_dict.update(dict(zip(__A , range(4 , num_words + 4 ) ) ) )
return vocab_dict
@torch.no_grad()
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Dict , __A : Any , __A : List[Any] , __A : int , __A : str , ):
'''simple docstring'''
snake_case: Union[str, Any] = WavaVecaConfig.from_pretrained(__A )
snake_case: str = SpeechaTextaConfig.from_pretrained(
__A , vocab_size=__A , decoder_layers=__A , do_stable_layer_norm=__A )
snake_case: List[str] = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=1_60_00 , padding_value=0 , do_normalize=__A , return_attention_mask=__A , )
snake_case , snake_case , snake_case: List[str] = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
snake_case: List[Any] = model[0].eval()
# set weights for wav2vec2 encoder
snake_case: Optional[Any] = WavaVecaModel(__A )
snake_case: Any = recursively_load_weights_wavaveca(model.encoder , __A )
snake_case: Union[str, Any] = SpeechaTextaForCausalLM(__A )
snake_case , snake_case: Optional[Any] = hf_decoder.model.decoder.load_state_dict(model.decoder.state_dict() , strict=__A )
# set output linear layer
unexpected_keys.remove('embed_out' )
snake_case: str = nn.Parameter(model.decoder.embed_out.detach() )
# layer norm is init to identity matrix so leaving it is fine
logger.warning(f"""The following keys are missing when loading the decoder weights: {missing_keys}""" )
logger.warning(f"""The following keys are unexpected when loading the decoder weights: {unexpected_keys}""" )
snake_case: int = SpeechEncoderDecoderModel(encoder=__A , decoder=__A )
snake_case: List[Any] = False
# add projection layer
snake_case: Union[str, Any] = nn.Parameter(projection_layer.weight )
snake_case: Union[str, Any] = nn.Parameter(projection_layer.bias )
snake_case: List[Any] = create_vocab_dict(__A )
with open(os.path.join(__A , 'vocab.json' ) , 'w' ) as fp:
json.dump(__A , __A )
snake_case: Union[str, Any] = SpeechaTextaTokenizer(os.path.join(__A , 'vocab.json' ) )
tokenizer.save_pretrained(__A )
snake_case: Tuple = hf_wavavec.config.to_dict()
snake_case: int = tokenizer.pad_token_id
snake_case: Dict = tokenizer.bos_token_id
snake_case: Optional[int] = tokenizer.eos_token_id
snake_case: Dict = 'speech_to_text_2'
snake_case: Optional[Any] = 'wav2vec2'
snake_case: Tuple = SpeechEncoderDecoderConfig.from_dict(__A )
hf_wavavec.save_pretrained(__A )
feature_extractor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
parser.add_argument("--dict_path", default=None, type=str, help="Path to dict of fine-tuned model")
parser.add_argument(
"--encoder_config_path",
default="facebook/wav2vec2-large-lv60",
type=str,
help="Path to hf encoder wav2vec2 checkpoint config",
)
parser.add_argument(
"--decoder_config_path",
default="facebook/s2t-small-mustc-en-fr-st",
type=str,
help="Path to hf decoder s2t checkpoint config",
)
parser.add_argument("--vocab_size", default=10_224, type=int, help="Vocab size of decoder")
parser.add_argument("--num_decoder_layers", default=7, type=int, help="Number of decoder layers")
__UpperCAmelCase = parser.parse_args()
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.dict_path,
encoder_config_path=args.encoder_config_path,
decoder_config_path=args.decoder_config_path,
vocab_size=args.vocab_size,
num_decoder_layers=args.num_decoder_layers,
) | 692 |
'''simple docstring'''
import importlib
import os
import fsspec
import pytest
from fsspec import register_implementation
from fsspec.registry import _registry as _fsspec_registry
from datasets.filesystems import COMPRESSION_FILESYSTEMS, HfFileSystem, extract_path_from_uri, is_remote_filesystem
from .utils import require_lza, require_zstandard
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
assert "mock" in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert "mock" not in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'mock-s3-bucket'
snake_case: int = f"""s3://{mock_bucket}"""
snake_case: Any = extract_path_from_uri(__A )
assert dataset_path.startswith('s3://' ) is False
snake_case: Union[str, Any] = './local/path'
snake_case: Union[str, Any] = extract_path_from_uri(__A )
assert dataset_path == new_dataset_path
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
snake_case: List[str] = is_remote_filesystem(__A )
assert is_remote is True
snake_case: int = fsspec.filesystem('file' )
snake_case: int = is_remote_filesystem(__A )
assert is_remote is False
@pytest.mark.parametrize('compression_fs_class' , __A )
def lowerCAmelCase_ ( __A : Optional[int] , __A : int , __A : str , __A : Optional[Any] , __A : List[str] , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: Optional[Any] = {'gzip': gz_file, 'xz': xz_file, 'zstd': zstd_file, 'bz2': bza_file, 'lz4': lza_file}
snake_case: Optional[int] = input_paths[compression_fs_class.protocol]
if input_path is None:
snake_case: str = f"""for '{compression_fs_class.protocol}' compression protocol, """
if compression_fs_class.protocol == "lz4":
reason += require_lza.kwargs["reason"]
elif compression_fs_class.protocol == "zstd":
reason += require_zstandard.kwargs["reason"]
pytest.skip(__A )
snake_case: List[str] = fsspec.filesystem(compression_fs_class.protocol , fo=__A )
assert isinstance(__A , __A )
snake_case: Any = os.path.basename(__A )
snake_case: int = expected_filename[: expected_filename.rindex('.' )]
assert fs.glob('*' ) == [expected_filename]
with fs.open(__A , 'r' , encoding='utf-8' ) as f, open(__A , encoding='utf-8' ) as expected_file:
assert f.read() == expected_file.read()
@pytest.mark.parametrize('protocol' , ['zip', 'gzip'] )
def lowerCAmelCase_ ( __A : Any , __A : int , __A : int ):
'''simple docstring'''
snake_case: List[str] = {'zip': zip_jsonl_path, 'gzip': jsonl_gz_path}
snake_case: str = compressed_file_paths[protocol]
snake_case: Dict = 'dataset.jsonl'
snake_case: Optional[Any] = f"""{protocol}://{member_file_path}::{compressed_file_path}"""
snake_case , *snake_case: List[Any] = fsspec.get_fs_token_paths(__A )
assert fs.isfile(__A )
assert not fs.isfile('non_existing_' + member_file_path )
@pytest.mark.integration
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Union[str, Any] , __A : List[Any] ):
'''simple docstring'''
snake_case: Tuple = hf_api.dataset_info(__A , token=__A )
snake_case: List[str] = HfFileSystem(repo_info=__A , token=__A )
assert sorted(hffs.glob('*' ) ) == [".gitattributes", "data"]
assert hffs.isdir('data' )
assert hffs.isfile('.gitattributes' ) and hffs.isfile('data/text_data.txt' )
with open(__A ) as f:
assert hffs.open('data/text_data.txt' , 'r' ).read() == f.read()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'bz2'
# Import module
import datasets.filesystems
# Overwrite protocol and reload
register_implementation(__A , __A , clobber=__A )
with pytest.warns(__A ) as warning_info:
importlib.reload(datasets.filesystems )
assert len(__A ) == 1
assert (
str(warning_info[0].message )
== f"""A filesystem protocol was already set for {protocol} and will be overwritten."""
) | 692 | 1 |
'''simple docstring'''
import argparse
import gc
import json
import os
import re
import torch
from huggingface_hub import hf_hub_download
from transformers import AutoModelForCausalLM, AutoTokenizer, PreTrainedTokenizerFast, RwkvConfig
from transformers.modeling_utils import WEIGHTS_INDEX_NAME, shard_checkpoint
__UpperCAmelCase = {
"169M": 12,
"430M": 24,
"1B5": 24,
"3B": 32,
"7B": 32,
"14B": 40,
}
__UpperCAmelCase = {
"169M": 768,
"430M": 1_024,
"1B5": 2_048,
"3B": 2_560,
"7B": 4_096,
"14B": 5_120,
}
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = list(state_dict.keys() )
for name in state_dict_keys:
snake_case: str = state_dict.pop(__A )
# emb -> embedding
if name.startswith('emb.' ):
snake_case: Dict = name.replace('emb.' , 'embeddings.' )
# ln_0 -> pre_ln (only present at block 0)
if name.startswith('blocks.0.ln0' ):
snake_case: Union[str, Any] = name.replace('blocks.0.ln0' , 'blocks.0.pre_ln' )
# att -> attention
snake_case: Union[str, Any] = re.sub(r'blocks\.(\d+)\.att' , r'blocks.\1.attention' , __A )
# ffn -> feed_forward
snake_case: Dict = re.sub(r'blocks\.(\d+)\.ffn' , r'blocks.\1.feed_forward' , __A )
# time_mix_k -> time_mix_key and reshape
if name.endswith('.time_mix_k' ):
snake_case: Union[str, Any] = name.replace('.time_mix_k' , '.time_mix_key' )
# time_mix_v -> time_mix_value and reshape
if name.endswith('.time_mix_v' ):
snake_case: List[Any] = name.replace('.time_mix_v' , '.time_mix_value' )
# time_mix_r -> time_mix_key and reshape
if name.endswith('.time_mix_r' ):
snake_case: List[Any] = name.replace('.time_mix_r' , '.time_mix_receptance' )
if name != "head.weight":
snake_case: Union[str, Any] = 'rwkv.' + name
snake_case: Optional[int] = weight
return state_dict
def lowerCAmelCase_ ( __A : Tuple , __A : Union[str, Any] , __A : Optional[Any] , __A : Optional[Any]=None , __A : str=None , __A : int=False , __A : Any=None ):
'''simple docstring'''
if tokenizer_file is None:
print('No `--tokenizer_file` provided, we will use the default tokenizer.' )
snake_case: Union[str, Any] = 5_02_77
snake_case: List[str] = AutoTokenizer.from_pretrained('EleutherAI/gpt-neox-20b' )
else:
snake_case: str = PreTrainedTokenizerFast(tokenizer_file=__A )
snake_case: Optional[int] = len(__A )
tokenizer.save_pretrained(__A )
# 2. Build the config
snake_case: int = list(NUM_HIDDEN_LAYERS_MAPPING.keys() )
if size is None:
# Try to infer size from the checkpoint name
for candidate in possible_sizes:
if candidate in checkpoint_file:
snake_case: List[str] = candidate
break
if size is None:
raise ValueError('Could not infer the size, please provide it with the `--size` argument.' )
if size not in possible_sizes:
raise ValueError(f"""`size` should be one of {possible_sizes}, got {size}.""" )
snake_case: str = RwkvConfig(
vocab_size=__A , num_hidden_layers=NUM_HIDDEN_LAYERS_MAPPING[size] , hidden_size=HIDEN_SIZE_MAPPING[size] , )
config.save_pretrained(__A )
# 3. Download model file then convert state_dict
snake_case: int = hf_hub_download(__A , __A )
snake_case: Any = torch.load(__A , map_location='cpu' )
snake_case: str = convert_state_dict(__A )
# 4. Split in shards and save
snake_case , snake_case: Optional[Any] = shard_checkpoint(__A )
for shard_file, shard in shards.items():
torch.save(__A , os.path.join(__A , __A ) )
if index is not None:
snake_case: int = os.path.join(__A , __A )
# Save the index as well
with open(__A , 'w' , encoding='utf-8' ) as f:
snake_case: List[str] = json.dumps(__A , indent=2 , sort_keys=__A ) + '\n'
f.write(__A )
# 5. Clean up shards (for some reason the file PyTorch saves take the same space as the whole state_dict
print(
'Cleaning up shards. This may error with an OOM error, it this is the case don\'t worry you still have converted the model.' )
snake_case: Any = list(shards.keys() )
del state_dict
del shards
gc.collect()
for shard_file in shard_files:
snake_case: Optional[Any] = torch.load(os.path.join(__A , __A ) )
torch.save({k: v.cpu().clone() for k, v in state_dict.items()} , os.path.join(__A , __A ) )
del state_dict
gc.collect()
if push_to_hub:
if model_name is None:
raise ValueError('Please provide a `model_name` to push the model to the Hub.' )
snake_case: str = AutoModelForCausalLM.from_pretrained(__A )
model.push_to_hub(__A , max_shard_size='2GB' )
tokenizer.push_to_hub(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--repo_id", default=None, type=str, required=True, help="Repo ID from which to pull the checkpoint."
)
parser.add_argument(
"--checkpoint_file", default=None, type=str, required=True, help="Name of the checkpoint file in the repo."
)
parser.add_argument(
"--output_dir", default=None, type=str, required=True, help="Where to save the converted model."
)
parser.add_argument(
"--tokenizer_file",
default=None,
type=str,
help="Path to the tokenizer file to use (if not provided, only the model is converted).",
)
parser.add_argument(
"--size",
default=None,
type=str,
help="Size of the model. Will be inferred from the `checkpoint_file` if not passed.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Push to the Hub the converted model.",
)
parser.add_argument(
"--model_name",
default=None,
type=str,
help="Name of the pushed model on the Hub, including the username / organization.",
)
__UpperCAmelCase = parser.parse_args()
convert_rmkv_checkpoint_to_hf_format(
args.repo_id,
args.checkpoint_file,
args.output_dir,
size=args.size,
tokenizer_file=args.tokenizer_file,
push_to_hub=args.push_to_hub,
model_name=args.model_name,
) | 692 |
'''simple docstring'''
import logging
import os
import random
import sys
from dataclasses import dataclass, field
from typing import Optional
import datasets
import numpy as np
import pandas as pd
from datasets import load_dataset
import transformers
from transformers import (
AutoConfig,
BartForSequenceClassification,
DataCollatorWithPadding,
EvalPrediction,
HfArgumentParser,
TapexTokenizer,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.17.0.dev0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
__UpperCAmelCase = logging.getLogger(__name__)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The name of the dataset to use (via the datasets library)."} )
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} , )
__UpperCamelCase = field(
default=1024 , metadata={
"help": (
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Overwrite the cached preprocessed datasets or not."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Whether to pad all samples to `max_seq_length`. "
"If False, will pad the samples dynamically when batching to the maximum length in the batch."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of prediction examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the training data."} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the validation data."} )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A csv or a json file containing the test data."} )
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.dataset_name is not None:
pass
elif self.train_file is None or self.validation_file is None:
raise ValueError('Need either a GLUE task, a training/validation file or a dataset name.' )
else:
snake_case: str = self.train_file.split('.' )[-1]
assert train_extension in ["csv", "json"], "`train_file` should be a csv or a json file."
snake_case: Optional[Any] = self.validation_file.split('.' )[-1]
assert (
validation_extension == train_extension
), "`validation_file` should have the same extension (csv or json) as `train_file`."
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."} , )
__UpperCamelCase = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case , snake_case , snake_case: Union[str, Any] = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case , snake_case , snake_case: str = parser.parse_args_into_dataclasses()
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
snake_case: Tuple = training_args.get_process_log_level()
logger.setLevel(__A )
datasets.utils.logging.set_verbosity(__A )
transformers.utils.logging.set_verbosity(__A )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"""
+ f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" )
logger.info(f"""Training/evaluation parameters {training_args}""" )
# Detecting last checkpoint.
snake_case: Any = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case: List[Any] = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"""Output directory ({training_args.output_dir}) already exists and is not empty. """
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Set seed before initializing model.
set_seed(training_args.seed )
# Get the datasets: you can either provide your own CSV/JSON training and evaluation files (see below)
# or specify a GLUE benchmark task (the dataset will be downloaded automatically from the datasets Hub).
#
# For JSON files, this script will use the `question` column for the input question and `table` column for the corresponding table.
#
# If the CSVs/JSONs contain only one non-label column, the script does single sentence classification on this
# single column. You can easily tweak this behavior (see below)
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
snake_case: int = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from your local files.
# CSV/JSON training and evaluation files are needed.
snake_case: Optional[int] = {'train': data_args.train_file, 'validation': data_args.validation_file}
# Get the test dataset: you can provide your own CSV/JSON test file (see below)
# when you use `do_predict` without specifying a GLUE benchmark task.
if training_args.do_predict:
if data_args.test_file is not None:
snake_case: Tuple = data_args.train_file.split('.' )[-1]
snake_case: Union[str, Any] = data_args.test_file.split('.' )[-1]
assert (
test_extension == train_extension
), "`test_file` should have the same extension (csv or json) as `train_file`."
snake_case: Union[str, Any] = data_args.test_file
else:
raise ValueError('Need either a GLUE task or a test file for `do_predict`.' )
for key in data_files.keys():
logger.info(f"""load a local file for {key}: {data_files[key]}""" )
if data_args.train_file.endswith('.csv' ):
# Loading a dataset from local csv files
snake_case: List[Any] = load_dataset('csv' , data_files=__A , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from local json files
snake_case: Optional[Any] = load_dataset('json' , data_files=__A , cache_dir=model_args.cache_dir )
# See more about loading any type of standard or custom dataset at
# https://huggingface.co/docs/datasets/loading_datasets.html.
# Labels
snake_case: Tuple = raw_datasets['train'].features['label'].names
snake_case: List[str] = len(__A )
# Load pretrained model and tokenizer
#
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case: Tuple = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# load tapex tokenizer
snake_case: List[str] = TapexTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , add_prefix_space=__A , )
snake_case: Union[str, Any] = BartForSequenceClassification.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# Padding strategy
if data_args.pad_to_max_length:
snake_case: int = 'max_length'
else:
# We will pad later, dynamically at batch creation, to the max sequence length in each batch
snake_case: Union[str, Any] = False
# Some models have set the order of the labels to use, so let's make sure we do use it.
snake_case: Optional[Any] = {'Refused': 0, 'Entailed': 1}
snake_case: List[Any] = {0: 'Refused', 1: 'Entailed'}
if data_args.max_seq_length > tokenizer.model_max_length:
logger.warning(
f"""The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"""
f"""model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}.""" )
snake_case: List[str] = min(data_args.max_seq_length , tokenizer.model_max_length )
def preprocess_tabfact_function(__A : Any ):
# Tokenize the texts
def _convert_table_text_to_pandas(__A : Dict ):
snake_case: str = [_table_row.split('#' ) for _table_row in _table_text.strip('\n' ).split('\n' )]
snake_case: List[str] = pd.DataFrame.from_records(_table_content[1:] , columns=_table_content[0] )
return _table_pd
snake_case: str = examples['statement']
snake_case: int = list(map(_convert_table_text_to_pandas , examples['table_text'] ) )
snake_case: List[Any] = tokenizer(__A , __A , padding=__A , max_length=__A , truncation=__A )
snake_case: List[Any] = examples['label']
return result
with training_args.main_process_first(desc='dataset map pre-processing' ):
snake_case: int = raw_datasets.map(
__A , batched=__A , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on dataset' , )
if training_args.do_train:
if "train" not in raw_datasets:
raise ValueError('--do_train requires a train dataset' )
snake_case: List[str] = raw_datasets['train']
if data_args.max_train_samples is not None:
snake_case: Tuple = train_dataset.select(range(data_args.max_train_samples ) )
if training_args.do_eval:
if "validation" not in raw_datasets and "validation_matched" not in raw_datasets:
raise ValueError('--do_eval requires a validation dataset' )
snake_case: Any = raw_datasets['validation']
if data_args.max_eval_samples is not None:
snake_case: Optional[int] = eval_dataset.select(range(data_args.max_eval_samples ) )
if training_args.do_predict or data_args.test_file is not None:
if "test" not in raw_datasets and "test_matched" not in raw_datasets:
raise ValueError('--do_predict requires a test dataset' )
snake_case: str = raw_datasets['test']
if data_args.max_predict_samples is not None:
snake_case: List[str] = predict_dataset.select(range(data_args.max_predict_samples ) )
# Log a few random samples from the training set:
if training_args.do_train:
for index in random.sample(range(len(__A ) ) , 3 ):
logger.info(f"""Sample {index} of the training set: {train_dataset[index]}.""" )
# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def compute_metrics(__A : EvalPrediction ):
snake_case: int = p.predictions[0] if isinstance(p.predictions , __A ) else p.predictions
snake_case: List[str] = np.argmax(__A , axis=1 )
return {"accuracy": (preds == p.label_ids).astype(np.floataa ).mean().item()}
# Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
if data_args.pad_to_max_length:
snake_case: str = default_data_collator
elif training_args.fpaa:
snake_case: List[str] = DataCollatorWithPadding(__A , pad_to_multiple_of=8 )
else:
snake_case: List[Any] = None
# Initialize our Trainer
snake_case: List[str] = Trainer(
model=__A , args=__A , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , compute_metrics=__A , tokenizer=__A , data_collator=__A , )
# Training
if training_args.do_train:
snake_case: Optional[int] = None
if training_args.resume_from_checkpoint is not None:
snake_case: str = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
snake_case: Optional[Any] = last_checkpoint
snake_case: Union[str, Any] = trainer.train(resume_from_checkpoint=__A )
snake_case: List[Any] = train_result.metrics
snake_case: List[Any] = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(__A )
)
snake_case: Optional[Any] = min(__A , len(__A ) )
trainer.save_model() # Saves the tokenizer too for easy upload
trainer.log_metrics('train' , __A )
trainer.save_metrics('train' , __A )
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info('*** Evaluate ***' )
snake_case: Dict = trainer.evaluate(eval_dataset=__A )
snake_case: Optional[int] = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(__A )
snake_case: Dict = min(__A , len(__A ) )
trainer.log_metrics('eval' , __A )
trainer.save_metrics('eval' , __A )
if training_args.do_predict:
logger.info('*** Predict ***' )
# Removing the `label` columns because it contains -1 and Trainer won't like that.
snake_case: Optional[int] = predict_dataset.remove_columns('label' )
snake_case: str = trainer.predict(__A , metric_key_prefix='predict' ).predictions
snake_case: Any = np.argmax(__A , axis=1 )
snake_case: int = os.path.join(training_args.output_dir , 'predict_results_tabfact.txt' )
if trainer.is_world_process_zero():
with open(__A , 'w' ) as writer:
logger.info('***** Predict Results *****' )
writer.write('index\tprediction\n' )
for index, item in enumerate(__A ):
snake_case: int = label_list[item]
writer.write(f"""{index}\t{item}\n""" )
snake_case: Optional[int] = {'finetuned_from': model_args.model_name_or_path, 'tasks': 'text-classification'}
if training_args.push_to_hub:
trainer.push_to_hub(**__A )
else:
trainer.create_model_card(**__A )
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
main()
if __name__ == "__main__":
main() | 692 | 1 |
'''simple docstring'''
import math
import unittest
from transformers import BioGptConfig, is_torch_available
from transformers.testing_utils import require_torch, slow, torch_device
from ...generation.test_utils import GenerationTesterMixin
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import (
BioGptForCausalLM,
BioGptForSequenceClassification,
BioGptForTokenClassification,
BioGptModel,
BioGptTokenizer,
)
from transformers.models.biogpt.modeling_biogpt import BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=13 , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=99 , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=5 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=37 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=None , ):
'''simple docstring'''
snake_case: List[str] = parent
snake_case: int = batch_size
snake_case: Any = seq_length
snake_case: Optional[int] = is_training
snake_case: List[Any] = use_input_mask
snake_case: Union[str, Any] = use_token_type_ids
snake_case: str = use_labels
snake_case: Optional[int] = vocab_size
snake_case: Tuple = hidden_size
snake_case: Dict = num_hidden_layers
snake_case: Optional[Any] = num_attention_heads
snake_case: Optional[Any] = intermediate_size
snake_case: Optional[int] = hidden_act
snake_case: int = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: Optional[int] = max_position_embeddings
snake_case: List[str] = type_vocab_size
snake_case: Union[str, Any] = type_sequence_label_size
snake_case: Dict = initializer_range
snake_case: Dict = num_labels
snake_case: Any = num_choices
snake_case: str = scope
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: List[str] = None
if self.use_input_mask:
snake_case: Dict = random_attention_mask([self.batch_size, self.seq_length] )
snake_case: Optional[int] = None
if self.use_token_type_ids:
snake_case: Optional[int] = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case: int = None
snake_case: Dict = None
snake_case: List[Any] = None
if self.use_labels:
snake_case: Optional[Any] = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case: Optional[int] = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
snake_case: Union[str, Any] = ids_tensor([self.batch_size] , self.num_choices )
snake_case: Optional[int] = self.get_config()
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def _UpperCamelCase ( self ):
'''simple docstring'''
return BioGptConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=SCREAMING_SNAKE_CASE__ , initializer_range=self.initializer_range , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = BioGptModel(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: List[str] = BioGptForCausalLM(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = BioGptModel(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
# create attention mask
snake_case: Any = torch.ones(input_ids.shape , dtype=torch.long , device=SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.seq_length // 2
snake_case: List[str] = 0
# first forward pass
snake_case , snake_case: Any = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ ).to_tuple()
# create hypothetical next token and extent to next_input_ids
snake_case: Union[str, Any] = ids_tensor((self.batch_size, 1) , config.vocab_size )
# change a random masked slice from input_ids
snake_case: Optional[int] = ids_tensor((1,) , SCREAMING_SNAKE_CASE__ ).item() + 1
snake_case: Optional[Any] = ids_tensor((self.batch_size, 1) , config.vocab_size ).squeeze(-1 )
snake_case: str = random_other_next_tokens
# append to next input_ids and attn_mask
snake_case: Any = torch.cat([input_ids, next_tokens] , dim=-1 )
snake_case: Dict = torch.cat(
[attn_mask, torch.ones((attn_mask.shape[0], 1) , dtype=torch.long , device=SCREAMING_SNAKE_CASE__ )] , dim=1 , )
# get two different outputs
snake_case: Optional[int] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )['last_hidden_state']
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , past_key_values=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )['last_hidden_state']
# select random slice
snake_case: Any = ids_tensor((1,) , output_from_past.shape[-1] ).item()
snake_case: Any = output_from_no_past[:, -1, random_slice_idx].detach()
snake_case: Optional[int] = output_from_past[:, 0, random_slice_idx].detach()
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-3 ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = BioGptModel(config=SCREAMING_SNAKE_CASE__ ).to(SCREAMING_SNAKE_CASE__ ).eval()
snake_case: Union[str, Any] = torch.ones(input_ids.shape , dtype=torch.long , device=SCREAMING_SNAKE_CASE__ )
# first forward pass
snake_case: List[str] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , use_cache=SCREAMING_SNAKE_CASE__ )
snake_case , snake_case: Dict = outputs.to_tuple()
# create hypothetical multiple next token and extent to next_input_ids
snake_case: Dict = ids_tensor((self.batch_size, 3) , config.vocab_size )
snake_case: int = ids_tensor((self.batch_size, 3) , 2 )
# append to next input_ids and
snake_case: Any = torch.cat([input_ids, next_tokens] , dim=-1 )
snake_case: Optional[int] = torch.cat([attention_mask, next_attn_mask] , dim=-1 )
snake_case: Optional[int] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )['last_hidden_state']
snake_case: Optional[int] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , past_key_values=SCREAMING_SNAKE_CASE__ )[
'last_hidden_state'
]
# select random slice
snake_case: str = ids_tensor((1,) , output_from_past.shape[-1] ).item()
snake_case: Optional[int] = output_from_no_past[:, -3:, random_slice_idx].detach()
snake_case: List[str] = output_from_past[:, :, random_slice_idx].detach()
self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1] )
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , atol=1E-3 ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
snake_case: str = BioGptForCausalLM(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
if gradient_checkpointing:
model.gradient_checkpointing_enable()
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.loss.shape , () )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
result.loss.backward()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Any = BioGptModel(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = model.config.initializer_range / math.sqrt(2 * model.config.num_hidden_layers )
for key in model.state_dict().keys():
if "c_proj" in key and "weight" in key:
self.parent.assertLessEqual(abs(torch.std(model.state_dict()[key] ) - model_std ) , 0.0_01 )
self.parent.assertLessEqual(abs(torch.mean(model.state_dict()[key] ) - 0.0 ) , 0.01 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = self.num_labels
snake_case: Any = BioGptForTokenClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.prepare_config_and_inputs()
(
(
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) ,
): List[str] = config_and_inputs
snake_case: int = {'input_ids': input_ids, 'attention_mask': input_mask}
return config, inputs_dict
@require_torch
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (
(BioGptModel, BioGptForCausalLM, BioGptForSequenceClassification, BioGptForTokenClassification)
if is_torch_available()
else ()
)
__UpperCamelCase = (BioGptForCausalLM,) if is_torch_available() else ()
__UpperCamelCase = (
{
"feature-extraction": BioGptModel,
"text-classification": BioGptForSequenceClassification,
"text-generation": BioGptForCausalLM,
"token-classification": BioGptForTokenClassification,
"zero-shot": BioGptForSequenceClassification,
}
if is_torch_available()
else {}
)
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = BioGptModelTester(self )
snake_case: Optional[int] = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , hidden_size=37 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.model_tester.prepare_config_and_inputs()
for type in ["absolute", "relative_key", "relative_key_query"]:
snake_case: Optional[Any] = type
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_model_attention_mask_past(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_forward_and_backwards(*SCREAMING_SNAKE_CASE__ , gradient_checkpointing=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_model_past_large_inputs(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_weight_initialization(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_for_token_classification(*SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = BioGptForCausalLM.from_pretrained('microsoft/biogpt' )
model.to(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = BioGptTokenizer.from_pretrained('microsoft/biogpt' )
snake_case: Optional[int] = 'left'
# Define PAD Token = EOS Token = 50256
snake_case: Optional[Any] = tokenizer.eos_token
snake_case: Optional[Any] = model.config.eos_token_id
# use different length sentences to test batching
snake_case: Tuple = [
'Hello, my dog is a little',
'Today, I',
]
snake_case: int = tokenizer(SCREAMING_SNAKE_CASE__ , return_tensors='pt' , padding=SCREAMING_SNAKE_CASE__ )
snake_case: str = inputs['input_ids'].to(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = model.generate(
input_ids=SCREAMING_SNAKE_CASE__ , attention_mask=inputs['attention_mask'].to(SCREAMING_SNAKE_CASE__ ) , )
snake_case: Optional[int] = tokenizer(sentences[0] , return_tensors='pt' ).input_ids.to(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = model.generate(input_ids=SCREAMING_SNAKE_CASE__ )
snake_case: int = inputs_non_padded.shape[-1] - inputs['attention_mask'][-1].long().sum().cpu().item()
snake_case: Tuple = tokenizer(sentences[1] , return_tensors='pt' ).input_ids.to(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = model.generate(input_ids=SCREAMING_SNAKE_CASE__ , max_length=model.config.max_length - num_paddings )
snake_case: Union[str, Any] = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.decode(output_non_padded[0] , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.decode(output_padded[0] , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = [
'Hello, my dog is a little bit bigger than a little bit.',
'Today, I have a good idea of how to use the information',
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , [non_padded_sentence, padded_sentence] )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
for model_name in BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: str = BioGptModel.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Dict = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: str = 3
snake_case: List[str] = input_dict['input_ids']
snake_case: Any = input_ids.ne(1 ).to(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size )
snake_case: Optional[Any] = BioGptForSequenceClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: int = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Dict = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[Any] = 3
snake_case: int = 'multi_label_classification'
snake_case: Dict = input_dict['input_ids']
snake_case: List[Any] = input_ids.ne(1 ).to(SCREAMING_SNAKE_CASE__ )
snake_case: int = ids_tensor(
[self.model_tester.batch_size, config.num_labels] , self.model_tester.type_sequence_label_size ).to(torch.float )
snake_case: int = BioGptForSequenceClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: str = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) )
@require_torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = BioGptForCausalLM.from_pretrained('microsoft/biogpt' )
snake_case: Any = torch.tensor([[2, 48_05, 9, 6_56, 21]] )
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ )[0]
snake_case: List[str] = 4_23_84
snake_case: Any = torch.Size((1, 5, vocab_size) )
self.assertEqual(output.shape , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = torch.tensor(
[[[-9.52_36, -9.89_18, 10.45_57], [-11.04_69, -9.64_23, 8.10_22], [-8.86_64, -7.88_26, 5.53_25]]] )
self.assertTrue(torch.allclose(output[:, :3, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = BioGptTokenizer.from_pretrained('microsoft/biogpt' )
snake_case: Optional[Any] = BioGptForCausalLM.from_pretrained('microsoft/biogpt' )
model.to(SCREAMING_SNAKE_CASE__ )
torch.manual_seed(0 )
snake_case: Optional[int] = tokenizer('COVID-19 is' , return_tensors='pt' ).to(SCREAMING_SNAKE_CASE__ )
snake_case: int = model.generate(
**SCREAMING_SNAKE_CASE__ , min_length=1_00 , max_length=10_24 , num_beams=5 , early_stopping=SCREAMING_SNAKE_CASE__ , )
snake_case: str = tokenizer.decode(output_ids[0] , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = (
'COVID-19 is a global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the'
' causative agent of coronavirus disease 2019 (COVID-19), which has spread to more than 200 countries and'
' territories, including the United States (US), Canada, Australia, New Zealand, the United Kingdom (UK),'
' and the United States of America (USA), as of March 11, 2020, with more than 800,000 confirmed cases and'
' more than 800,000 deaths.'
)
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
import math
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or number % 2 == 0 or number % 3 == 0:
# Negatives, 0, 1, all even numbers, all multiples of 3 are not primes
return False
# All primes number are in format of 6k +/- 1
for i in range(5 , int(math.sqrt(__A ) + 1 ) , 6 ):
if number % i == 0 or number % (i + 2) == 0:
return False
return True
def lowerCAmelCase_ ( __A : float = 0.1 ):
'''simple docstring'''
snake_case: Optional[int] = 3
snake_case: int = 3
while primes / (2 * j - 1) >= ratio:
for i in range(j * j + j + 1 , (j + 2) * (j + 2) , j + 1 ):
primes += is_prime(__A )
j += 2
return j
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available
__UpperCAmelCase = {}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["MLukeTokenizer"]
if TYPE_CHECKING:
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mluke import MLukeTokenizer
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCAmelCase = "pt"
elif is_tf_available():
__UpperCAmelCase = "tf"
else:
__UpperCAmelCase = "jax"
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ByTaTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: int = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ByTaTokenizer.from_pretrained('google/byt5-small' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=20 , SCREAMING_SNAKE_CASE__=5 ):
'''simple docstring'''
snake_case: Optional[Any] = []
for i in range(len(SCREAMING_SNAKE_CASE__ ) ):
try:
snake_case: Optional[Any] = tokenizer.decode([i] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
except UnicodeDecodeError:
pass
toks.append((i, tok) )
snake_case: List[str] = list(filter(lambda SCREAMING_SNAKE_CASE__ : re.match(r'^[ a-zA-Z]+$' , t[1] ) , SCREAMING_SNAKE_CASE__ ) )
snake_case: str = list(filter(lambda SCREAMING_SNAKE_CASE__ : [t[0]] == tokenizer.encode(t[1] , add_special_tokens=SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ ) )
if max_length is not None and len(SCREAMING_SNAKE_CASE__ ) > max_length:
snake_case: Union[str, Any] = toks[:max_length]
if min_length is not None and len(SCREAMING_SNAKE_CASE__ ) < min_length and len(SCREAMING_SNAKE_CASE__ ) > 0:
while len(SCREAMING_SNAKE_CASE__ ) < min_length:
snake_case: Tuple = toks + toks
# toks_str = [t[1] for t in toks]
snake_case: Dict = [t[0] for t in toks]
# Ensure consistency
snake_case: int = tokenizer.decode(SCREAMING_SNAKE_CASE__ , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
if " " not in output_txt and len(SCREAMING_SNAKE_CASE__ ) > 1:
snake_case: str = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
)
if with_prefix_space:
snake_case: Tuple = ' ' + output_txt
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
return output_txt, output_ids
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: str = tokenizer(['hi</s>', 'I went to the gym</s>', '</s>'] )
snake_case: List[Any] = tokenizer(['hi', 'I went to the gym', ''] )
self.assertListEqual(batch_with_eos_added['input_ids'] , batch_without_eos_added['input_ids'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: Union[str, Any] = 'Unicode €.'
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [88, 1_13, 1_08, 1_02, 1_14, 1_03, 1_04, 35, 2_29, 1_33, 1_75, 49, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[str] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'Unicode €.</s>' )
snake_case: List[Any] = tokenizer('e è é ê ë' )
snake_case: Optional[Any] = [1_04, 35, 1_98, 1_71, 35, 1_98, 1_72, 35, 1_98, 1_73, 35, 1_98, 1_74, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[Any] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'e è é ê ë</s>' )
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë' ) ) , 'e è é ê ë</s>' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.ta_base_tokenizer
snake_case: Optional[Any] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case: Optional[int] = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 1, 0]
# fmt: on
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if FRAMEWORK != "jax":
snake_case: Optional[Any] = list(batch.input_ids.numpy()[0] )
else:
snake_case: Dict = list(batch.input_ids.tolist()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 37) , batch.input_ids.shape )
self.assertEqual((2, 37) , batch.attention_mask.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.ta_base_tokenizer
snake_case: List[str] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case: Union[str, Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_attention_mask' , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.ta_base_tokenizer
snake_case: str = [
'Summary of the text.',
'Another summary.',
]
snake_case: Dict = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.ta_base_tokenizer
snake_case: Optional[int] = ['A long paragraph for summarization. </s>']
snake_case: str = ['Summary of the text. </s>']
# fmt: off
snake_case: str = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 35, 1]
snake_case: Optional[int] = [86, 1_20, 1_12, 1_12, 1_00, 1_17, 1_24, 35, 1_14, 1_05, 35, 1_19, 1_07, 1_04, 35, 1_19, 1_04, 1_23, 1_19, 49, 35, 1]
# fmt: on
snake_case: List[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , text_target=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['input_ids'][0] )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['labels'][0] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: Optional[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: Union[str, Any] = tempfile.mkdtemp()
snake_case: Dict = ' He is very happy, UNwant\u00E9d,running'
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'] )
snake_case: List[str] = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token' )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: int = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn('new_additional_special_token' , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Union[str, Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: str = json.load(SCREAMING_SNAKE_CASE__ )
snake_case: int = [F"""<extra_id_{i}>""" for i in range(1_25 )]
snake_case: Optional[int] = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case: str = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Dict = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , )
self.assertIn(
'an_additional_special_token' , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['an_additional_special_token'] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'] ) ) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: Union[str, Any] = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token' , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: Union[str, Any] = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , )
self.assertIn('a_new_additional_special_token' , tokenizer.additional_special_tokens )
self.assertEqual(
['a_new_additional_special_token'] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'] ) ) , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertTrue(tokenizer.decode([2_55] ) == '' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_tokenizers(fast=SCREAMING_SNAKE_CASE__ , do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = ['t', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 'x', 't', '</s>']
snake_case: List[str] = tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Optional[Any] = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Dict = 0
snake_case: List[Any] = tokenizer.convert_ids_to_tokens(
SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [token_id_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [token_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [token_id_to_test_setters] ) | 692 | 1 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 |
'''simple docstring'''
from typing import Any, Dict, Optional
import torch
import torch.nn.functional as F
from torch import nn
from ..utils import maybe_allow_in_graph
from .activations import get_activation
from .attention_processor import Attention
from .embeddings import CombinedTimestepLabelEmbeddings
@maybe_allow_in_graph
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "layer_norm" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = only_cross_attention
snake_case: Optional[Any] = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm_zero'
snake_case: Tuple = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm'
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
raise ValueError(
F"""`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"""
F""" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}.""" )
# Define 3 blocks. Each block has its own normalization layer.
# 1. Self-Attn
if self.use_ada_layer_norm:
snake_case: List[str] = AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case: str = AdaLayerNormZero(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
else:
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if only_cross_attention else None , upcast_attention=SCREAMING_SNAKE_CASE__ , )
# 2. Cross-Attn
if cross_attention_dim is not None or double_self_attention:
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
# the second cross attention block.
snake_case: Tuple = (
AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm
else nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if not double_self_attention else None , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , upcast_attention=SCREAMING_SNAKE_CASE__ , ) # is self-attn if encoder_hidden_states is none
else:
snake_case: int = None
snake_case: Tuple = None
# 3. Feed-forward
snake_case: Union[str, Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = FeedForward(SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , activation_fn=SCREAMING_SNAKE_CASE__ , final_dropout=SCREAMING_SNAKE_CASE__ )
# let chunk size default to None
snake_case: Any = None
snake_case: Any = 0
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = chunk_size
snake_case: str = dim
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
if self.use_ada_layer_norm:
snake_case: Optional[int] = self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case , snake_case , snake_case , snake_case , snake_case: int = self.norma(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=hidden_states.dtype )
else:
snake_case: List[str] = self.norma(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = cross_attention_kwargs if cross_attention_kwargs is not None else {}
snake_case: List[str] = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=encoder_hidden_states if self.only_cross_attention else None , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if self.use_ada_layer_norm_zero:
snake_case: Tuple = gate_msa.unsqueeze(1 ) * attn_output
snake_case: List[str] = attn_output + hidden_states
# 2. Cross-Attention
if self.attna is not None:
snake_case: Dict = (
self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) if self.use_ada_layer_norm else self.norma(SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: List[str] = attn_output + hidden_states
# 3. Feed-forward
snake_case: str = self.norma(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: str = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
if self._chunk_size is not None:
# "feed_forward_chunk_size" can be used to save memory
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
raise ValueError(
F"""`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`.""" )
snake_case: List[str] = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
snake_case: Optional[Any] = torch.cat(
[self.ff(SCREAMING_SNAKE_CASE__ ) for hid_slice in norm_hidden_states.chunk(SCREAMING_SNAKE_CASE__ , dim=self._chunk_dim )] , dim=self._chunk_dim , )
else:
snake_case: int = self.ff(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: Union[str, Any] = gate_mlp.unsqueeze(1 ) * ff_output
snake_case: Tuple = ff_output + hidden_states
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 4 , SCREAMING_SNAKE_CASE__ = 0.0 , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: int = int(dim * mult )
snake_case: Optional[Any] = dim_out if dim_out is not None else dim
if activation_fn == "gelu":
snake_case: int = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if activation_fn == "gelu-approximate":
snake_case: Optional[Any] = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , approximate='tanh' )
elif activation_fn == "geglu":
snake_case: List[Any] = GEGLU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif activation_fn == "geglu-approximate":
snake_case: Optional[int] = ApproximateGELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.ModuleList([] )
# project in
self.net.append(SCREAMING_SNAKE_CASE__ )
# project dropout
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
# project out
self.net.append(nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
if final_dropout:
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for module in self.net:
snake_case: Optional[int] = module(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "none" ):
'''simple docstring'''
super().__init__()
snake_case: Optional[int] = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = approximate
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ , approximate=self.approximate )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) , approximate=self.approximate ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.proj(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.gelu(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = nn.Linear(SCREAMING_SNAKE_CASE__ , dim_out * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: int = self.proj(SCREAMING_SNAKE_CASE__ ).chunk(2 , dim=-1 )
return hidden_states * self.gelu(SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = self.proj(SCREAMING_SNAKE_CASE__ )
return x * torch.sigmoid(1.7_02 * x )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Optional[Any] = nn.Embedding(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = nn.SiLU()
snake_case: Union[str, Any] = nn.Linear(SCREAMING_SNAKE_CASE__ , embedding_dim * 2 )
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case: Dict = torch.chunk(SCREAMING_SNAKE_CASE__ , 2 )
snake_case: str = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale) + shift
return x
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = CombinedTimestepLabelEmbeddings(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.SiLU()
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , 6 * embedding_dim , bias=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ , eps=1E-6 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
snake_case: int = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case , snake_case , snake_case , snake_case , snake_case: str = emb.chunk(6 , dim=1 )
snake_case: Dict = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale_msa[:, None]) + shift_msa[:, None]
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 1E-5 ):
'''simple docstring'''
super().__init__()
snake_case: str = num_groups
snake_case: str = eps
if act_fn is None:
snake_case: Dict = None
else:
snake_case: List[str] = get_activation(SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , out_dim * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.act:
snake_case: Optional[Any] = self.act(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.linear(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = emb[:, :, None, None]
snake_case , snake_case: List[Any] = emb.chunk(2 , dim=1 )
snake_case: Any = F.group_norm(SCREAMING_SNAKE_CASE__ , self.num_groups , eps=self.eps )
snake_case: Optional[int] = x * (1 + scale) + shift
return x | 692 | 1 |
'''simple docstring'''
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/beit-base-patch16-224-pt22k": (
"https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json"
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "beit"
def __init__( self , SCREAMING_SNAKE_CASE__=81_92 , SCREAMING_SNAKE_CASE__=7_68 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=30_72 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-12 , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=[3, 5, 7, 11] , SCREAMING_SNAKE_CASE__=[1, 2, 3, 6] , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=0.4 , SCREAMING_SNAKE_CASE__=2_56 , SCREAMING_SNAKE_CASE__=1 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=2_55 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = vocab_size
snake_case: int = hidden_size
snake_case: int = num_hidden_layers
snake_case: str = num_attention_heads
snake_case: str = intermediate_size
snake_case: Union[str, Any] = hidden_act
snake_case: List[Any] = hidden_dropout_prob
snake_case: Any = attention_probs_dropout_prob
snake_case: Union[str, Any] = initializer_range
snake_case: List[Any] = layer_norm_eps
snake_case: Tuple = image_size
snake_case: Dict = patch_size
snake_case: List[Any] = num_channels
snake_case: List[str] = use_mask_token
snake_case: Optional[Any] = use_absolute_position_embeddings
snake_case: str = use_relative_position_bias
snake_case: List[str] = use_shared_relative_position_bias
snake_case: Tuple = layer_scale_init_value
snake_case: Dict = drop_path_rate
snake_case: Union[str, Any] = use_mean_pooling
# decode head attributes (semantic segmentation)
snake_case: Any = out_indices
snake_case: Union[str, Any] = pool_scales
# auxiliary head attributes (semantic segmentation)
snake_case: Optional[Any] = use_auxiliary_head
snake_case: int = auxiliary_loss_weight
snake_case: Union[str, Any] = auxiliary_channels
snake_case: Union[str, Any] = auxiliary_num_convs
snake_case: Optional[int] = auxiliary_concat_input
snake_case: Dict = semantic_loss_ignore_index
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = version.parse("1.11" )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 1E-4 | 692 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
# limitations under the License.
# NOTE: This file is deprecated and will be removed in a future version.
# It only exists so that temporarely `from diffusers.pipelines import DiffusionPipeline` works
from .pipelines import DiffusionPipeline, ImagePipelineOutput # noqa: F401
from .utils import deprecate
deprecate(
"pipelines_utils",
"0.22.0",
"Importing `DiffusionPipeline` or `ImagePipelineOutput` from diffusers.pipeline_utils is deprecated. Please import from diffusers.pipelines.pipeline_utils instead.",
standard_warn=False,
stacklevel=3,
) | 692 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available
__UpperCAmelCase = {
"configuration_biogpt": ["BIOGPT_PRETRAINED_CONFIG_ARCHIVE_MAP", "BioGptConfig"],
"tokenization_biogpt": ["BioGptTokenizer"],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST",
"BioGptForCausalLM",
"BioGptForTokenClassification",
"BioGptForSequenceClassification",
"BioGptModel",
"BioGptPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_biogpt import BIOGPT_PRETRAINED_CONFIG_ARCHIVE_MAP, BioGptConfig
from .tokenization_biogpt import BioGptTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_biogpt import (
BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST,
BioGptForCausalLM,
BioGptForSequenceClassification,
BioGptForTokenClassification,
BioGptModel,
BioGptPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_roformer": ["ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoFormerConfig", "RoFormerOnnxConfig"],
"tokenization_roformer": ["RoFormerTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["RoFormerTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoFormerForCausalLM",
"RoFormerForMaskedLM",
"RoFormerForMultipleChoice",
"RoFormerForQuestionAnswering",
"RoFormerForSequenceClassification",
"RoFormerForTokenClassification",
"RoFormerLayer",
"RoFormerModel",
"RoFormerPreTrainedModel",
"load_tf_weights_in_roformer",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFRoFormerForCausalLM",
"TFRoFormerForMaskedLM",
"TFRoFormerForMultipleChoice",
"TFRoFormerForQuestionAnswering",
"TFRoFormerForSequenceClassification",
"TFRoFormerForTokenClassification",
"TFRoFormerLayer",
"TFRoFormerModel",
"TFRoFormerPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"FlaxRoFormerForMaskedLM",
"FlaxRoFormerForMultipleChoice",
"FlaxRoFormerForQuestionAnswering",
"FlaxRoFormerForSequenceClassification",
"FlaxRoFormerForTokenClassification",
"FlaxRoFormerModel",
"FlaxRoFormerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig
from .tokenization_roformer import RoFormerTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_roformer_fast import RoFormerTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roformer import (
ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
RoFormerForCausalLM,
RoFormerForMaskedLM,
RoFormerForMultipleChoice,
RoFormerForQuestionAnswering,
RoFormerForSequenceClassification,
RoFormerForTokenClassification,
RoFormerLayer,
RoFormerModel,
RoFormerPreTrainedModel,
load_tf_weights_in_roformer,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_roformer import (
TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TFRoFormerForCausalLM,
TFRoFormerForMaskedLM,
TFRoFormerForMultipleChoice,
TFRoFormerForQuestionAnswering,
TFRoFormerForSequenceClassification,
TFRoFormerForTokenClassification,
TFRoFormerLayer,
TFRoFormerModel,
TFRoFormerPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_roformer import (
FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
FlaxRoFormerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 | 1 |
'''simple docstring'''
import unittest
from transformers import MraConfig, is_torch_available
from transformers.testing_utils import require_torch, slow, torch_device
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
if is_torch_available():
import torch
from transformers import (
MraForMaskedLM,
MraForMultipleChoice,
MraForQuestionAnswering,
MraForSequenceClassification,
MraForTokenClassification,
MraModel,
)
from transformers.models.mra.modeling_mra import MRA_PRETRAINED_MODEL_ARCHIVE_LIST
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=8 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=99 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=5 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=36 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=None , ):
'''simple docstring'''
snake_case: List[str] = parent
snake_case: Any = batch_size
snake_case: str = seq_length
snake_case: Tuple = is_training
snake_case: Optional[int] = use_input_mask
snake_case: List[str] = use_token_type_ids
snake_case: Optional[int] = use_labels
snake_case: List[Any] = vocab_size
snake_case: Union[str, Any] = hidden_size
snake_case: int = num_hidden_layers
snake_case: List[str] = num_attention_heads
snake_case: Any = intermediate_size
snake_case: Tuple = hidden_act
snake_case: Tuple = hidden_dropout_prob
snake_case: Optional[int] = attention_probs_dropout_prob
snake_case: List[Any] = max_position_embeddings
snake_case: Union[str, Any] = type_vocab_size
snake_case: Any = type_sequence_label_size
snake_case: Dict = initializer_range
snake_case: Any = num_labels
snake_case: str = num_choices
snake_case: List[str] = scope
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: Tuple = None
if self.use_input_mask:
snake_case: List[str] = random_attention_mask([self.batch_size, self.seq_length] )
snake_case: Dict = None
if self.use_token_type_ids:
snake_case: Optional[Any] = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case: Any = None
snake_case: Any = None
snake_case: List[str] = None
if self.use_labels:
snake_case: List[Any] = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case: Any = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
snake_case: Any = ids_tensor([self.batch_size] , self.num_choices )
snake_case: Optional[int] = self.get_config()
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def _UpperCamelCase ( self ):
'''simple docstring'''
return MraConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=SCREAMING_SNAKE_CASE__ , initializer_range=self.initializer_range , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_config()
snake_case: Any = 3_00
return config
def _UpperCamelCase ( self ):
'''simple docstring'''
(
(
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) ,
): Optional[Any] = self.prepare_config_and_inputs()
snake_case: List[Any] = True
snake_case: Tuple = floats_tensor([self.batch_size, self.seq_length, self.hidden_size] )
snake_case: str = ids_tensor([self.batch_size, self.seq_length] , vocab_size=2 )
return (
config,
input_ids,
token_type_ids,
input_mask,
sequence_labels,
token_labels,
choice_labels,
encoder_hidden_states,
encoder_attention_mask,
)
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = MraModel(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: str = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = model(SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: int = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: List[str] = True
snake_case: Optional[Any] = MraModel(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: str = model(
SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , encoder_attention_mask=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = model(
SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , )
snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = MraForMaskedLM(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = MraForQuestionAnswering(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Any = model(
SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , start_positions=SCREAMING_SNAKE_CASE__ , end_positions=SCREAMING_SNAKE_CASE__ , )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = self.num_labels
snake_case: Tuple = MraForSequenceClassification(SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Optional[int] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: str = self.num_labels
snake_case: List[Any] = MraForTokenClassification(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Any = self.num_choices
snake_case: Tuple = MraForMultipleChoice(config=SCREAMING_SNAKE_CASE__ )
model.to(SCREAMING_SNAKE_CASE__ )
model.eval()
snake_case: Optional[Any] = input_ids.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous()
snake_case: Any = token_type_ids.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous()
snake_case: Any = input_mask.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous()
snake_case: str = model(
SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ , labels=SCREAMING_SNAKE_CASE__ , )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_choices) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.prepare_config_and_inputs()
(
(
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) ,
): Union[str, Any] = config_and_inputs
snake_case: Any = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask': input_mask}
return config, inputs_dict
@require_torch
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (
(
MraModel,
MraForMaskedLM,
MraForMultipleChoice,
MraForQuestionAnswering,
MraForSequenceClassification,
MraForTokenClassification,
)
if is_torch_available()
else ()
)
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = ()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = MraModelTester(self )
snake_case: List[Any] = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , hidden_size=37 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.model_tester.prepare_config_and_inputs()
for type in ["absolute", "relative_key", "relative_key_query"]:
snake_case: int = type
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_masked_lm(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_multiple_choice(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_question_answering(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_sequence_classification(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_token_classification(*SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
for model_name in MRA_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: Tuple = MraModel.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
@unittest.skip(reason='MRA does not output attentions' )
def _UpperCamelCase ( self ):
'''simple docstring'''
return
@require_torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = MraModel.from_pretrained('uw-madison/mra-base-512-4' )
snake_case: Dict = torch.arange(2_56 ).unsqueeze(0 )
with torch.no_grad():
snake_case: Any = model(SCREAMING_SNAKE_CASE__ )[0]
snake_case: Dict = torch.Size((1, 2_56, 7_68) )
self.assertEqual(output.shape , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = torch.tensor(
[[[-0.01_40, 0.08_30, -0.03_81], [0.15_46, 0.14_02, 0.02_20], [0.11_62, 0.08_51, 0.01_65]]] )
self.assertTrue(torch.allclose(output[:, :3, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = MraForMaskedLM.from_pretrained('uw-madison/mra-base-512-4' )
snake_case: Optional[Any] = torch.arange(2_56 ).unsqueeze(0 )
with torch.no_grad():
snake_case: int = model(SCREAMING_SNAKE_CASE__ )[0]
snake_case: Dict = 5_02_65
snake_case: int = torch.Size((1, 2_56, vocab_size) )
self.assertEqual(output.shape , SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = torch.tensor(
[[[9.25_95, -3.60_38, 11.88_19], [9.38_69, -3.26_93, 11.09_56], [11.85_24, -3.49_38, 13.12_10]]] )
self.assertTrue(torch.allclose(output[:, :3, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = MraForMaskedLM.from_pretrained('uw-madison/mra-base-4096-8-d3' )
snake_case: int = torch.arange(40_96 ).unsqueeze(0 )
with torch.no_grad():
snake_case: Dict = model(SCREAMING_SNAKE_CASE__ )[0]
snake_case: Union[str, Any] = 5_02_65
snake_case: Optional[Any] = torch.Size((1, 40_96, vocab_size) )
self.assertEqual(output.shape , SCREAMING_SNAKE_CASE__ )
snake_case: str = torch.tensor(
[[[5.47_89, -2.35_64, 7.50_64], [7.90_67, -1.33_69, 9.96_68], [9.07_12, -1.81_06, 7.03_80]]] )
self.assertTrue(torch.allclose(output[:, :3, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) ) | 692 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 | 1 |
'''simple docstring'''
import inspect
import unittest
from transformers import RegNetConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from transformers.utils import cached_property, is_vision_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_flax_common import FlaxModelTesterMixin, floats_tensor
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.models.regnet.modeling_flax_regnet import FlaxRegNetForImageClassification, FlaxRegNetModel
if is_vision_available():
from PIL import Image
from transformers import AutoImageProcessor
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=10 , SCREAMING_SNAKE_CASE__=[10, 20, 30, 40] , SCREAMING_SNAKE_CASE__=[1, 1, 2, 1] , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__="relu" , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=None , ):
'''simple docstring'''
snake_case: Dict = parent
snake_case: Dict = batch_size
snake_case: Optional[Any] = image_size
snake_case: Tuple = num_channels
snake_case: Optional[Any] = embeddings_size
snake_case: int = hidden_sizes
snake_case: List[str] = depths
snake_case: Optional[int] = is_training
snake_case: Union[str, Any] = use_labels
snake_case: int = hidden_act
snake_case: Tuple = num_labels
snake_case: List[str] = scope
snake_case: Union[str, Any] = len(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
snake_case: Tuple = self.get_config()
return config, pixel_values
def _UpperCamelCase ( self ):
'''simple docstring'''
return RegNetConfig(
num_channels=self.num_channels , embeddings_size=self.embeddings_size , hidden_sizes=self.hidden_sizes , depths=self.depths , hidden_act=self.hidden_act , num_labels=self.num_labels , image_size=self.image_size , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = FlaxRegNetModel(config=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ )
# Output shape (b, c, h, w)
self.parent.assertEqual(
result.last_hidden_state.shape , (self.batch_size, self.hidden_sizes[-1], self.image_size // 32, self.image_size // 32) , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Tuple = self.num_labels
snake_case: List[Any] = FlaxRegNetForImageClassification(config=SCREAMING_SNAKE_CASE__ )
snake_case: str = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.prepare_config_and_inputs()
snake_case , snake_case: Union[str, Any] = config_and_inputs
snake_case: Optional[int] = {'pixel_values': pixel_values}
return config, inputs_dict
@require_flax
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (FlaxRegNetModel, FlaxRegNetForImageClassification) if is_flax_available() else ()
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = FlaxRegNetModelTester(self )
snake_case: Optional[int] = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , has_text_modality=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.create_and_test_config_common_properties()
self.config_tester.create_and_test_config_to_json_string()
self.config_tester.create_and_test_config_to_json_file()
self.config_tester.create_and_test_config_from_and_save_pretrained()
self.config_tester.create_and_test_config_with_num_labels()
self.config_tester.check_config_can_be_init_without_params()
self.config_tester.check_config_arguments_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_image_classification(*SCREAMING_SNAKE_CASE__ )
@unittest.skip(reason='RegNet does not use inputs_embeds' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@unittest.skip(reason='RegNet does not support input and output embeddings' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case: str = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = inspect.signature(model.__call__ )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
snake_case: Any = [*signature.parameters.keys()]
snake_case: List[Any] = ['pixel_values']
self.assertListEqual(arg_names[:1] , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
def check_hidden_states_output(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: str = model(**self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
snake_case: List[Any] = outputs.encoder_hidden_states if config.is_encoder_decoder else outputs.hidden_states
snake_case: Optional[Any] = self.model_tester.num_stages
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , expected_num_stages + 1 )
snake_case , snake_case: Dict = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case: Union[str, Any] = True
check_hidden_states_output(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# check that output_hidden_states also work using config
del inputs_dict["output_hidden_states"]
snake_case: Optional[int] = True
check_hidden_states_output(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Tuple = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
snake_case: Union[str, Any] = self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = model_class(SCREAMING_SNAKE_CASE__ )
@jax.jit
def model_jitted(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
return model(pixel_values=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
with self.subTest('JIT Enabled' ):
snake_case: List[Any] = model_jitted(**SCREAMING_SNAKE_CASE__ ).to_tuple()
with self.subTest('JIT Disabled' ):
with jax.disable_jit():
snake_case: Any = model_jitted(**SCREAMING_SNAKE_CASE__ ).to_tuple()
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , len(SCREAMING_SNAKE_CASE__ ) )
for jitted_output, output in zip(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
self.assertEqual(jitted_output.shape , output.shape )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: List[Any] = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' )
return image
@require_flax
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return AutoImageProcessor.from_pretrained('facebook/regnet-y-040' ) if is_vision_available() else None
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = FlaxRegNetForImageClassification.from_pretrained('facebook/regnet-y-040' )
snake_case: Union[str, Any] = self.default_image_processor
snake_case: List[str] = prepare_img()
snake_case: Optional[Any] = image_processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='np' )
snake_case: Any = model(**SCREAMING_SNAKE_CASE__ )
# verify the logits
snake_case: Any = (1, 10_00)
self.assertEqual(outputs.logits.shape , SCREAMING_SNAKE_CASE__ )
snake_case: int = jnp.array([-0.41_80, -1.50_51, -3.48_36] )
self.assertTrue(jnp.allclose(outputs.logits[0, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) ) | 692 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 | 1 |
'''simple docstring'''
import os
import unittest
from transformers.models.transfo_xl.tokenization_transfo_xl import VOCAB_FILES_NAMES, TransfoXLTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = TransfoXLTokenizer
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Dict = [
'<unk>',
'[CLS]',
'[SEP]',
'want',
'unwanted',
'wa',
'un',
'running',
',',
'low',
'l',
]
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = True
return TransfoXLTokenizer.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: str = '<unk> UNwanted , running'
snake_case: int = '<unk> unwanted, running'
return input_text, output_text
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = TransfoXLTokenizer(vocab_file=self.vocab_file , lower_case=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = tokenizer.tokenize('<unk> UNwanted , running' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['<unk>', 'unwanted', ',', 'running'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [0, 4, 8, 7] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = TransfoXLTokenizer(lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo ! how \n Are yoU ? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = TransfoXLTokenizer(lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo ! how \n Are yoU ? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = TransfoXLTokenizer(lower_case=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = 'Hello (bracket) and side-scrolled [and] Henry\'s $5,000 with 3.34 m. What\'s up!?'
snake_case: Optional[int] = [
'Hello',
'(',
'bracket',
')',
'and',
'side',
'@-@',
'scrolled',
'[',
'and',
']',
'Henry',
'\'s',
'$',
'5',
'@,@',
'000',
'with',
'3',
'@.@',
'34',
'm',
'.',
'What',
'\'s',
'up',
'!',
'?',
]
self.assertListEqual(tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizer()
snake_case: List[Any] = len(SCREAMING_SNAKE_CASE__ )
tokenizer.add_tokens(['new1', 'new2'] )
tokenizer.move_added_token('new1' , 1 )
# Check that moved token is not copied (duplicate)
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , original_len + 2 )
# Check that token is moved to specified id
self.assertEqual(tokenizer.encode('new1' ) , [1] )
self.assertEqual(tokenizer.decode([1] ) , 'new1' ) | 692 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 | 1 |
'''simple docstring'''
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = len(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [0] * len_array
if len_array > 0:
snake_case: Optional[int] = array[0]
for i in range(1 , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = self.prefix_sum[i - 1] + array[i]
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if start == 0:
return self.prefix_sum[end]
return self.prefix_sum[end] - self.prefix_sum[start - 1]
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = {0}
for sum_item in self.prefix_sum:
if sum_item - target_sum in sums:
return True
sums.add(SCREAMING_SNAKE_CASE__ )
return False
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from collections import defaultdict
from typing import Optional
from ..image_utils import load_image
from ..utils import (
add_end_docstrings,
is_torch_available,
logging,
requires_backends,
)
from .base import PIPELINE_INIT_ARGS, ChunkPipeline
if is_torch_available():
import torch
from ..models.auto.modeling_auto import MODEL_FOR_MASK_GENERATION_MAPPING
__UpperCAmelCase = logging.get_logger(__name__)
@add_end_docstrings(snake_case )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
requires_backends(self , 'vision' )
requires_backends(self , 'torch' )
if self.framework != "pt":
raise ValueError(F"""The {self.__class__} is only available in PyTorch.""" )
self.check_model_type(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Any = {}
snake_case: int = {}
snake_case: Tuple = {}
# preprocess args
if "points_per_batch" in kwargs:
snake_case: Optional[int] = kwargs['points_per_batch']
if "points_per_crop" in kwargs:
snake_case: Any = kwargs['points_per_crop']
if "crops_n_layers" in kwargs:
snake_case: Union[str, Any] = kwargs['crops_n_layers']
if "crop_overlap_ratio" in kwargs:
snake_case: List[str] = kwargs['crop_overlap_ratio']
if "crop_n_points_downscale_factor" in kwargs:
snake_case: Optional[Any] = kwargs['crop_n_points_downscale_factor']
# postprocess args
if "pred_iou_thresh" in kwargs:
snake_case: Tuple = kwargs['pred_iou_thresh']
if "stability_score_offset" in kwargs:
snake_case: List[Any] = kwargs['stability_score_offset']
if "mask_threshold" in kwargs:
snake_case: Tuple = kwargs['mask_threshold']
if "stability_score_thresh" in kwargs:
snake_case: str = kwargs['stability_score_thresh']
if "crops_nms_thresh" in kwargs:
snake_case: Tuple = kwargs['crops_nms_thresh']
if "output_rle_mask" in kwargs:
snake_case: str = kwargs['output_rle_mask']
if "output_bboxes_mask" in kwargs:
snake_case: int = kwargs['output_bboxes_mask']
return preprocess_kwargs, forward_params, postprocess_kwargs
def __call__( self , SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=None , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return super().__call__(SCREAMING_SNAKE_CASE__ , *SCREAMING_SNAKE_CASE__ , num_workers=SCREAMING_SNAKE_CASE__ , batch_size=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=64 , SCREAMING_SNAKE_CASE__ = 0 , SCREAMING_SNAKE_CASE__ = 5_12 / 15_00 , SCREAMING_SNAKE_CASE__ = 32 , SCREAMING_SNAKE_CASE__ = 1 , ):
'''simple docstring'''
snake_case: Tuple = load_image(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.image_processor.size['longest_edge']
snake_case , snake_case , snake_case , snake_case: str = self.image_processor.generate_crop_boxes(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.image_processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='pt' )
with self.device_placement():
if self.framework == "pt":
snake_case: Any = self.get_inference_context()
with inference_context():
snake_case: Optional[Any] = self._ensure_tensor_on_device(SCREAMING_SNAKE_CASE__ , device=self.device )
snake_case: Any = self.model.get_image_embeddings(model_inputs.pop('pixel_values' ) )
snake_case: Optional[int] = image_embeddings
snake_case: str = grid_points.shape[1]
snake_case: Union[str, Any] = points_per_batch if points_per_batch is not None else n_points
if points_per_batch <= 0:
raise ValueError(
'Cannot have points_per_batch<=0. Must be >=1 to returned batched outputs. '
'To return all points at once, set points_per_batch to None' )
for i in range(0 , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: Dict = grid_points[:, i : i + points_per_batch, :, :]
snake_case: Any = input_labels[:, i : i + points_per_batch]
snake_case: Optional[int] = i == n_points - points_per_batch
yield {
"input_points": batched_points,
"input_labels": labels,
"input_boxes": crop_boxes,
"is_last": is_last,
**model_inputs,
}
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.88 , SCREAMING_SNAKE_CASE__=0.95 , SCREAMING_SNAKE_CASE__=0 , SCREAMING_SNAKE_CASE__=1 , ):
'''simple docstring'''
snake_case: Dict = model_inputs.pop('input_boxes' )
snake_case: Optional[Any] = model_inputs.pop('is_last' )
snake_case: Optional[Any] = model_inputs.pop('original_sizes' ).tolist()
snake_case: Tuple = model_inputs.pop('reshaped_input_sizes' ).tolist()
snake_case: Any = self.model(**SCREAMING_SNAKE_CASE__ )
# post processing happens here in order to avoid CPU GPU copies of ALL the masks
snake_case: Optional[Any] = model_outputs['pred_masks']
snake_case: str = self.image_processor.post_process_masks(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , binarize=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = model_outputs['iou_scores']
snake_case , snake_case , snake_case: str = self.image_processor.filter_masks(
masks[0] , iou_scores[0] , original_sizes[0] , input_boxes[0] , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , )
return {
"masks": masks,
"is_last": is_last,
"boxes": boxes,
"iou_scores": iou_scores,
}
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.7 , ):
'''simple docstring'''
snake_case: Union[str, Any] = []
snake_case: List[Any] = []
snake_case: Optional[Any] = []
for model_output in model_outputs:
all_scores.append(model_output.pop('iou_scores' ) )
all_masks.extend(model_output.pop('masks' ) )
all_boxes.append(model_output.pop('boxes' ) )
snake_case: List[str] = torch.cat(SCREAMING_SNAKE_CASE__ )
snake_case: int = torch.cat(SCREAMING_SNAKE_CASE__ )
snake_case , snake_case , snake_case , snake_case: Dict = self.image_processor.post_process_for_mask_generation(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = defaultdict(SCREAMING_SNAKE_CASE__ )
for output in model_outputs:
for k, v in output.items():
extra[k].append(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = {}
if output_rle_mask:
snake_case: int = rle_mask
if output_bboxes_mask:
snake_case: Optional[Any] = bounding_boxes
return {"masks": output_masks, "scores": iou_scores, **optional, **extra} | 692 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = "▁"
__UpperCAmelCase = {"vocab_file": "sentencepiece.bpe.model"}
__UpperCAmelCase = {
"vocab_file": {
"facebook/xglm-564M": "https://huggingface.co/facebook/xglm-564M/resolve/main/sentencepiece.bpe.model",
}
}
__UpperCAmelCase = {
"facebook/xglm-564M": 2_048,
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
# Compatibility with the original tokenizer
snake_case: Optional[Any] = 7
snake_case: List[str] = [F"""<madeupword{i}>""" for i in range(self.num_madeup_words )]
snake_case: str = kwargs.get('additional_special_tokens' , [] )
kwargs["additional_special_tokens"] += [
word for word in madeup_words if word not in kwargs["additional_special_tokens"]
]
super().__init__(
bos_token=SCREAMING_SNAKE_CASE__ , eos_token=SCREAMING_SNAKE_CASE__ , unk_token=SCREAMING_SNAKE_CASE__ , sep_token=SCREAMING_SNAKE_CASE__ , cls_token=SCREAMING_SNAKE_CASE__ , pad_token=SCREAMING_SNAKE_CASE__ , sp_model_kwargs=self.sp_model_kwargs , **SCREAMING_SNAKE_CASE__ , )
snake_case: int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(SCREAMING_SNAKE_CASE__ ) )
snake_case: int = vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
snake_case: Tuple = 1
# Mimic fairseq token-to-id alignment for the first 4 token
snake_case: Optional[Any] = {'<s>': 0, '<pad>': 1, '</s>': 2, '<unk>': 3}
snake_case: Union[str, Any] = len(self.sp_model )
snake_case: str = {F"""<madeupword{i}>""": sp_size + i + self.fairseq_offset for i in range(self.num_madeup_words )}
self.fairseq_tokens_to_ids.update(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def __getstate__( self ):
'''simple docstring'''
snake_case: List[Any] = self.__dict__.copy()
snake_case: Union[str, Any] = None
snake_case: Union[str, Any] = self.sp_model.serialized_model_proto()
return state
def __setstate__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
snake_case: Union[str, Any] = {}
snake_case: Tuple = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.LoadFromSerializedProto(self.sp_model_proto )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.sep_token_id] + token_ids_a
snake_case: Optional[Any] = [self.sep_token_id]
return sep + token_ids_a + sep + sep + token_ids_a
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=SCREAMING_SNAKE_CASE__ , token_ids_a=SCREAMING_SNAKE_CASE__ , already_has_special_tokens=SCREAMING_SNAKE_CASE__ )
if token_ids_a is None:
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ )) + [1, 1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
if token_ids_a is None:
return len(sep + token_ids_a ) * [0]
return len(sep + token_ids_a + sep + sep + token_ids_a ) * [0]
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.sp_model ) + self.fairseq_offset + self.num_madeup_words
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = {self.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.sp_model.encode(SCREAMING_SNAKE_CASE__ , out_type=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
snake_case: Dict = self.sp_model.PieceToId(SCREAMING_SNAKE_CASE__ )
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = ''.join(SCREAMING_SNAKE_CASE__ ).replace(SCREAMING_SNAKE_CASE__ , ' ' ).strip()
return out_string
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(SCREAMING_SNAKE_CASE__ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: List[str] = os.path.join(
SCREAMING_SNAKE_CASE__ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(SCREAMING_SNAKE_CASE__ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , SCREAMING_SNAKE_CASE__ )
elif not os.path.isfile(self.vocab_file ):
with open(SCREAMING_SNAKE_CASE__ , 'wb' ) as fi:
snake_case: int = self.sp_model.serialized_model_proto()
fi.write(SCREAMING_SNAKE_CASE__ )
return (out_vocab_file,) | 692 | 1 |
'''simple docstring'''
import sys
__UpperCAmelCase = (
"73167176531330624919225119674426574742355349194934"
"96983520312774506326239578318016984801869478851843"
"85861560789112949495459501737958331952853208805511"
"12540698747158523863050715693290963295227443043557"
"66896648950445244523161731856403098711121722383113"
"62229893423380308135336276614282806444486645238749"
"30358907296290491560440772390713810515859307960866"
"70172427121883998797908792274921901699720888093776"
"65727333001053367881220235421809751254540594752243"
"52584907711670556013604839586446706324415722155397"
"53697817977846174064955149290862569321978468622482"
"83972241375657056057490261407972968652414535100474"
"82166370484403199890008895243450658541227588666881"
"16427171479924442928230863465674813919123162824586"
"17866458359124566529476545682848912883142607690042"
"24219022671055626321111109370544217506941658960408"
"07198403850962455444362981230987879927244284909188"
"84580156166097919133875499200524063689912560717606"
"05886116467109405077541002256983155200055935729725"
"71636269561882670428252483600823257530420752963450"
)
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: Tuple = 1
for digit in s:
product *= int(__A )
return product
def lowerCAmelCase_ ( __A : str = N ):
'''simple docstring'''
snake_case: List[str] = -sys.maxsize - 1
snake_case: int = n[:13]
snake_case: Union[str, Any] = 13
while cur_index < len(__A ) - 13:
if int(n[cur_index] ) >= int(substr[0] ):
snake_case: Optional[Any] = substr[1:] + n[cur_index]
cur_index += 1
else:
snake_case: Dict = max(__A , str_eval(__A ) )
snake_case: Optional[int] = n[cur_index : cur_index + 13]
cur_index += 13
return largest_product
if __name__ == "__main__":
print(F'{solution() = }') | 692 |
'''simple docstring'''
from operator import delitem, getitem, setitem
import pytest
from data_structures.hashing.hash_map import HashMap
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
return getitem, k
def lowerCAmelCase_ ( __A : Any , __A : Optional[int] ):
'''simple docstring'''
return setitem, k, v
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
return delitem, k
def lowerCAmelCase_ ( __A : str , __A : int , *__A : Tuple ):
'''simple docstring'''
try:
return fun(__A , *__A ), None
except Exception as e:
return None, e
__UpperCAmelCase = (
_set("key_a", "val_a"),
_set("key_b", "val_b"),
)
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_a", "val_b"),
]
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_b", "val_b"),
_del("key_a"),
_del("key_b"),
_set("key_a", "val_a"),
_del("key_a"),
]
__UpperCAmelCase = [
_get("key_a"),
_del("key_a"),
_set("key_a", "val_a"),
_del("key_a"),
_del("key_a"),
_get("key_a"),
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
*[_del(x) for x in range(5)],
_set("key_a", "val_b"),
]
@pytest.mark.parametrize(
'operations' , (
pytest.param(_add_items , id='add items' ),
pytest.param(_overwrite_items , id='overwrite items' ),
pytest.param(_delete_items , id='delete items' ),
pytest.param(_access_absent_items , id='access absent items' ),
pytest.param(_add_with_resize_up , id='add with resize up' ),
pytest.param(_add_with_resize_down , id='add with resize down' ),
) , )
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[Any] = HashMap(initial_block_size=4 )
snake_case: List[Any] = {}
for _, (fun, *args) in enumerate(__A ):
snake_case , snake_case: Optional[int] = _run_operation(__A , __A , *__A )
snake_case , snake_case: str = _run_operation(__A , __A , *__A )
assert my_res == py_res
assert str(__A ) == str(__A )
assert set(__A ) == set(__A )
assert len(__A ) == len(__A )
assert set(my.items() ) == set(py.items() )
def lowerCAmelCase_ ( ):
'''simple docstring'''
def is_public(__A : str ) -> bool:
return not name.startswith('_' )
snake_case: Dict = {name for name in dir({} ) if is_public(__A )}
snake_case: List[str] = {name for name in dir(HashMap() ) if is_public(__A )}
assert dict_public_names > hash_public_names | 692 | 1 |
'''simple docstring'''
import gc
import unittest
import numpy as np
import torch
from torch.backends.cuda import sdp_kernel
from diffusers import (
CMStochasticIterativeScheduler,
ConsistencyModelPipeline,
UNetaDModel,
)
from diffusers.utils import randn_tensor, slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_a, require_torch_gpu
from ..pipeline_params import UNCONDITIONAL_IMAGE_GENERATION_BATCH_PARAMS, UNCONDITIONAL_IMAGE_GENERATION_PARAMS
from ..test_pipelines_common import PipelineTesterMixin
enable_full_determinism()
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ConsistencyModelPipeline
__UpperCamelCase = UNCONDITIONAL_IMAGE_GENERATION_PARAMS
__UpperCamelCase = UNCONDITIONAL_IMAGE_GENERATION_BATCH_PARAMS
# Override required_optional_params to remove num_images_per_prompt
__UpperCamelCase = frozenset(
[
"num_inference_steps",
"generator",
"latents",
"output_type",
"return_dict",
"callback",
"callback_steps",
] )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = UNetaDModel.from_pretrained(
'diffusers/consistency-models-test' , subfolder='test_unet' , )
return unet
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = UNetaDModel.from_pretrained(
'diffusers/consistency-models-test' , subfolder='test_unet_class_cond' , )
return unet
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if class_cond:
snake_case: Dict = self.dummy_cond_unet
else:
snake_case: List[Any] = self.dummy_uncond_unet
# Default to CM multistep sampler
snake_case: Optional[Any] = CMStochasticIterativeScheduler(
num_train_timesteps=40 , sigma_min=0.0_02 , sigma_max=80.0 , )
snake_case: Union[str, Any] = {
'unet': unet,
'scheduler': scheduler,
}
return components
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0 ):
'''simple docstring'''
if str(SCREAMING_SNAKE_CASE__ ).startswith('mps' ):
snake_case: Any = torch.manual_seed(SCREAMING_SNAKE_CASE__ )
else:
snake_case: Union[str, Any] = torch.Generator(device=SCREAMING_SNAKE_CASE__ ).manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: int = {
'batch_size': 1,
'num_inference_steps': None,
'timesteps': [22, 0],
'generator': generator,
'output_type': 'np',
}
return inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Union[str, Any] = self.get_dummy_components()
snake_case: Tuple = ConsistencyModelPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: Any = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: int = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 32, 32, 3)
snake_case: str = image[0, -3:, -3:, -1]
snake_case: List[str] = np.array([0.35_72, 0.62_73, 0.40_31, 0.39_61, 0.43_21, 0.57_30, 0.52_66, 0.47_80, 0.50_04] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Any = self.get_dummy_components(class_cond=SCREAMING_SNAKE_CASE__ )
snake_case: Any = ConsistencyModelPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = 0
snake_case: int = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 32, 32, 3)
snake_case: Optional[Any] = image[0, -3:, -3:, -1]
snake_case: Optional[int] = np.array([0.35_72, 0.62_73, 0.40_31, 0.39_61, 0.43_21, 0.57_30, 0.52_66, 0.47_80, 0.50_04] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: List[Any] = self.get_dummy_components()
snake_case: Dict = ConsistencyModelPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: str = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = 1
snake_case: Dict = None
snake_case: Union[str, Any] = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 32, 32, 3)
snake_case: str = image[0, -3:, -3:, -1]
snake_case: List[Any] = np.array([0.50_04, 0.50_04, 0.49_94, 0.50_08, 0.49_76, 0.50_18, 0.49_90, 0.49_82, 0.49_87] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = 'cpu' # ensure determinism for the device-dependent torch.Generator
snake_case: Optional[int] = self.get_dummy_components(class_cond=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = ConsistencyModelPipeline(**SCREAMING_SNAKE_CASE__ )
snake_case: int = pipe.to(SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.get_dummy_inputs(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = 1
snake_case: int = None
snake_case: Optional[int] = 0
snake_case: Any = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 32, 32, 3)
snake_case: List[str] = image[0, -3:, -3:, -1]
snake_case: Optional[int] = np.array([0.50_04, 0.50_04, 0.49_94, 0.50_08, 0.49_76, 0.50_18, 0.49_90, 0.49_82, 0.49_87] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
@slow
@require_torch_gpu
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__=0 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__="cpu" , SCREAMING_SNAKE_CASE__=torch.floataa , SCREAMING_SNAKE_CASE__=(1, 3, 64, 64) ):
'''simple docstring'''
snake_case: str = torch.manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = {
'num_inference_steps': None,
'timesteps': [22, 0],
'class_labels': 0,
'generator': generator,
'output_type': 'np',
}
if get_fixed_latents:
snake_case: Any = self.get_fixed_latents(seed=SCREAMING_SNAKE_CASE__ , device=SCREAMING_SNAKE_CASE__ , dtype=SCREAMING_SNAKE_CASE__ , shape=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = latents
return inputs
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__=0 , SCREAMING_SNAKE_CASE__="cpu" , SCREAMING_SNAKE_CASE__=torch.floataa , SCREAMING_SNAKE_CASE__=(1, 3, 64, 64) ):
'''simple docstring'''
if type(SCREAMING_SNAKE_CASE__ ) == str:
snake_case: Optional[int] = torch.device(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = torch.Generator(device=SCREAMING_SNAKE_CASE__ ).manual_seed(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = randn_tensor(SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ , device=SCREAMING_SNAKE_CASE__ , dtype=SCREAMING_SNAKE_CASE__ )
return latents
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = UNetaDModel.from_pretrained('diffusers/consistency_models' , subfolder='diffusers_cd_imagenet64_l2' )
snake_case: int = CMStochasticIterativeScheduler(
num_train_timesteps=40 , sigma_min=0.0_02 , sigma_max=80.0 , )
snake_case: Optional[int] = ConsistencyModelPipeline(unet=SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ )
pipe.to(torch_device=SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: str = self.get_inputs()
snake_case: Optional[int] = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 64, 64, 3)
snake_case: Tuple = image[0, -3:, -3:, -1]
snake_case: Optional[Any] = np.array([0.08_88, 0.08_81, 0.06_66, 0.04_79, 0.02_92, 0.01_95, 0.02_01, 0.01_63, 0.02_54] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 2E-2
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = UNetaDModel.from_pretrained('diffusers/consistency_models' , subfolder='diffusers_cd_imagenet64_l2' )
snake_case: int = CMStochasticIterativeScheduler(
num_train_timesteps=40 , sigma_min=0.0_02 , sigma_max=80.0 , )
snake_case: Optional[Any] = ConsistencyModelPipeline(unet=SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ )
pipe.to(torch_device=SCREAMING_SNAKE_CASE__ )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: str = self.get_inputs()
snake_case: Union[str, Any] = 1
snake_case: List[str] = None
snake_case: int = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 64, 64, 3)
snake_case: Optional[int] = image[0, -3:, -3:, -1]
snake_case: str = np.array([0.03_40, 0.01_52, 0.00_63, 0.02_67, 0.02_21, 0.01_07, 0.04_16, 0.01_86, 0.02_17] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 2E-2
@require_torch_a
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = UNetaDModel.from_pretrained('diffusers/consistency_models' , subfolder='diffusers_cd_imagenet64_l2' )
snake_case: Dict = CMStochasticIterativeScheduler(
num_train_timesteps=40 , sigma_min=0.0_02 , sigma_max=80.0 , )
snake_case: str = ConsistencyModelPipeline(unet=SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ )
pipe.to(torch_device=SCREAMING_SNAKE_CASE__ , torch_dtype=torch.floataa )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.get_inputs(get_fixed_latents=SCREAMING_SNAKE_CASE__ , device=SCREAMING_SNAKE_CASE__ )
# Ensure usage of flash attention in torch 2.0
with sdp_kernel(enable_flash=SCREAMING_SNAKE_CASE__ , enable_math=SCREAMING_SNAKE_CASE__ , enable_mem_efficient=SCREAMING_SNAKE_CASE__ ):
snake_case: List[str] = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 64, 64, 3)
snake_case: Union[str, Any] = image[0, -3:, -3:, -1]
snake_case: Dict = np.array([0.18_75, 0.14_28, 0.12_89, 0.21_51, 0.20_92, 0.14_77, 0.18_77, 0.16_41, 0.13_53] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
@require_torch_a
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = UNetaDModel.from_pretrained('diffusers/consistency_models' , subfolder='diffusers_cd_imagenet64_l2' )
snake_case: List[Any] = CMStochasticIterativeScheduler(
num_train_timesteps=40 , sigma_min=0.0_02 , sigma_max=80.0 , )
snake_case: Tuple = ConsistencyModelPipeline(unet=SCREAMING_SNAKE_CASE__ , scheduler=SCREAMING_SNAKE_CASE__ )
pipe.to(torch_device=SCREAMING_SNAKE_CASE__ , torch_dtype=torch.floataa )
pipe.set_progress_bar_config(disable=SCREAMING_SNAKE_CASE__ )
snake_case: str = self.get_inputs(get_fixed_latents=SCREAMING_SNAKE_CASE__ , device=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = 1
snake_case: Any = None
# Ensure usage of flash attention in torch 2.0
with sdp_kernel(enable_flash=SCREAMING_SNAKE_CASE__ , enable_math=SCREAMING_SNAKE_CASE__ , enable_mem_efficient=SCREAMING_SNAKE_CASE__ ):
snake_case: List[str] = pipe(**SCREAMING_SNAKE_CASE__ ).images
assert image.shape == (1, 64, 64, 3)
snake_case: List[str] = image[0, -3:, -3:, -1]
snake_case: Union[str, Any] = np.array([0.16_63, 0.19_48, 0.22_75, 0.16_80, 0.12_04, 0.12_45, 0.18_58, 0.13_38, 0.20_95] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3 | 692 |
'''simple docstring'''
import argparse
import json
import os
import fairseq
import torch
from torch import nn
from transformers import (
SpeechaTextaConfig,
SpeechaTextaForCausalLM,
SpeechaTextaTokenizer,
SpeechEncoderDecoderConfig,
SpeechEncoderDecoderModel,
WavaVecaConfig,
WavaVecaFeatureExtractor,
WavaVecaModel,
logging,
)
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"post_extract_proj": "feature_projection.projection",
"encoder.pos_conv.0": "encoder.pos_conv_embed.conv",
"self_attn.k_proj": "encoder.layers.*.attention.k_proj",
"self_attn.v_proj": "encoder.layers.*.attention.v_proj",
"self_attn.q_proj": "encoder.layers.*.attention.q_proj",
"self_attn.out_proj": "encoder.layers.*.attention.out_proj",
"self_attn_layer_norm": "encoder.layers.*.layer_norm",
"fc1": "encoder.layers.*.feed_forward.intermediate_dense",
"fc2": "encoder.layers.*.feed_forward.output_dense",
"final_layer_norm": "encoder.layers.*.final_layer_norm",
"encoder.layer_norm": "encoder.layer_norm",
"w2v_model.layer_norm": "feature_projection.layer_norm",
"quantizer.weight_proj": "quantizer.weight_proj",
"quantizer.vars": "quantizer.codevectors",
"project_q": "project_q",
"final_proj": "project_hid",
"w2v_encoder.proj": "lm_head",
"mask_emb": "masked_spec_embed",
}
__UpperCAmelCase = [
"lm_head",
"quantizer.weight_proj",
"quantizer.codevectors",
"project_q",
"project_hid",
]
def lowerCAmelCase_ ( __A : Any , __A : Optional[Any] , __A : Union[str, Any] , __A : int , __A : Optional[int] ):
'''simple docstring'''
for attribute in key.split('.' ):
snake_case: List[str] = getattr(__A , __A )
if weight_type is not None:
snake_case: Optional[int] = getattr(__A , __A ).shape
else:
snake_case: Optional[int] = hf_pointer.shape
assert hf_shape == value.shape, (
f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
snake_case: Optional[int] = value
elif weight_type == "weight_g":
snake_case: List[str] = value
elif weight_type == "weight_v":
snake_case: Dict = value
elif weight_type == "bias":
snake_case: Optional[Any] = value
else:
snake_case: int = value
logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def lowerCAmelCase_ ( __A : List[Any] , __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = []
snake_case: List[Any] = fairseq_model.state_dict()
snake_case: Union[str, Any] = hf_model.feature_extractor
# if encoder has different dim to decoder -> use proj_weight
snake_case: Dict = None
for name, value in fairseq_dict.items():
snake_case: Tuple = False
if "conv_layers" in name:
load_conv_layer(
__A , __A , __A , __A , hf_model.config.feat_extract_norm == 'group' , )
snake_case: List[Any] = True
elif name.split('.' )[0] == "proj":
snake_case: List[Any] = fairseq_model.proj
snake_case: int = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
snake_case: int = True
if "*" in mapped_key:
snake_case: List[str] = name.split(__A )[0].split('.' )[-2]
snake_case: Dict = mapped_key.replace('*' , __A )
if "weight_g" in name:
snake_case: Tuple = 'weight_g'
elif "weight_v" in name:
snake_case: int = 'weight_v'
elif "bias" in name:
snake_case: Tuple = 'bias'
elif "weight" in name:
snake_case: List[Any] = 'weight'
else:
snake_case: Any = None
set_recursively(__A , __A , __A , __A , __A )
continue
if not is_used:
unused_weights.append(__A )
logger.warning(f"""Unused weights: {unused_weights}""" )
return proj_weight
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : int , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = full_name.split('conv_layers.' )[-1]
snake_case: Tuple = name.split('.' )
snake_case: Any = int(items[0] )
snake_case: Optional[int] = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
snake_case: Tuple = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
snake_case: int = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
snake_case: Any = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
snake_case: str = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(__A )
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case , snake_case: List[Any] = emb.weight.shape
snake_case: Optional[int] = nn.Linear(__A , __A , bias=__A )
snake_case: Any = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
with open(__A , 'r' , encoding='utf-8' ) as f:
snake_case: List[Any] = f.readlines()
snake_case: Any = [line.split(' ' )[0] for line in lines]
snake_case: int = len(__A )
snake_case: Dict = {
'<s>': 0,
'<pad>': 1,
'</s>': 2,
'<unk>': 3,
}
vocab_dict.update(dict(zip(__A , range(4 , num_words + 4 ) ) ) )
return vocab_dict
@torch.no_grad()
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Dict , __A : Any , __A : List[Any] , __A : int , __A : str , ):
'''simple docstring'''
snake_case: Union[str, Any] = WavaVecaConfig.from_pretrained(__A )
snake_case: str = SpeechaTextaConfig.from_pretrained(
__A , vocab_size=__A , decoder_layers=__A , do_stable_layer_norm=__A )
snake_case: List[str] = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=1_60_00 , padding_value=0 , do_normalize=__A , return_attention_mask=__A , )
snake_case , snake_case , snake_case: List[str] = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
snake_case: List[Any] = model[0].eval()
# set weights for wav2vec2 encoder
snake_case: Optional[Any] = WavaVecaModel(__A )
snake_case: Any = recursively_load_weights_wavaveca(model.encoder , __A )
snake_case: Union[str, Any] = SpeechaTextaForCausalLM(__A )
snake_case , snake_case: Optional[Any] = hf_decoder.model.decoder.load_state_dict(model.decoder.state_dict() , strict=__A )
# set output linear layer
unexpected_keys.remove('embed_out' )
snake_case: str = nn.Parameter(model.decoder.embed_out.detach() )
# layer norm is init to identity matrix so leaving it is fine
logger.warning(f"""The following keys are missing when loading the decoder weights: {missing_keys}""" )
logger.warning(f"""The following keys are unexpected when loading the decoder weights: {unexpected_keys}""" )
snake_case: int = SpeechEncoderDecoderModel(encoder=__A , decoder=__A )
snake_case: List[Any] = False
# add projection layer
snake_case: Union[str, Any] = nn.Parameter(projection_layer.weight )
snake_case: Union[str, Any] = nn.Parameter(projection_layer.bias )
snake_case: List[Any] = create_vocab_dict(__A )
with open(os.path.join(__A , 'vocab.json' ) , 'w' ) as fp:
json.dump(__A , __A )
snake_case: Union[str, Any] = SpeechaTextaTokenizer(os.path.join(__A , 'vocab.json' ) )
tokenizer.save_pretrained(__A )
snake_case: Tuple = hf_wavavec.config.to_dict()
snake_case: int = tokenizer.pad_token_id
snake_case: Dict = tokenizer.bos_token_id
snake_case: Optional[int] = tokenizer.eos_token_id
snake_case: Dict = 'speech_to_text_2'
snake_case: Optional[Any] = 'wav2vec2'
snake_case: Tuple = SpeechEncoderDecoderConfig.from_dict(__A )
hf_wavavec.save_pretrained(__A )
feature_extractor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
parser.add_argument("--dict_path", default=None, type=str, help="Path to dict of fine-tuned model")
parser.add_argument(
"--encoder_config_path",
default="facebook/wav2vec2-large-lv60",
type=str,
help="Path to hf encoder wav2vec2 checkpoint config",
)
parser.add_argument(
"--decoder_config_path",
default="facebook/s2t-small-mustc-en-fr-st",
type=str,
help="Path to hf decoder s2t checkpoint config",
)
parser.add_argument("--vocab_size", default=10_224, type=int, help="Vocab size of decoder")
parser.add_argument("--num_decoder_layers", default=7, type=int, help="Number of decoder layers")
__UpperCAmelCase = parser.parse_args()
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.dict_path,
encoder_config_path=args.encoder_config_path,
decoder_config_path=args.decoder_config_path,
vocab_size=args.vocab_size,
num_decoder_layers=args.num_decoder_layers,
) | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = [
"DownloadConfig",
"DownloadManager",
"DownloadMode",
"StreamingDownloadManager",
]
from .download_config import DownloadConfig
from .download_manager import DownloadManager, DownloadMode
from .streaming_download_manager import StreamingDownloadManager | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 1_00 ):
'''simple docstring'''
snake_case: List[str] = n * (n + 1) * (2 * n + 1) / 6
snake_case: List[Any] = (n * (n + 1) / 2) ** 2
return int(square_of_sum - sum_of_squares )
if __name__ == "__main__":
print(F'{solution() = }') | 692 | 1 |
'''simple docstring'''
import doctest
import logging
import os
import unittest
from pathlib import Path
from typing import List, Union
import transformers
from transformers.testing_utils import require_tf, require_torch, slow
__UpperCAmelCase = logging.getLogger()
@unittest.skip("Temporarily disable the doc tests." )
@require_torch
@require_tf
@slow
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , ):
'''simple docstring'''
snake_case: Optional[Any] = [file for file in os.listdir(SCREAMING_SNAKE_CASE__ ) if os.path.isfile(os.path.join(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )]
if identifier is not None:
snake_case: Dict = [file for file in files if identifier in file]
if n_identifier is not None:
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
for n_ in n_identifier:
snake_case: List[Any] = [file for file in files if n_ not in file]
else:
snake_case: Tuple = [file for file in files if n_identifier not in file]
snake_case: Dict = ignore_files or []
ignore_files.append('__init__.py' )
snake_case: Tuple = [file for file in files if file not in ignore_files]
for file in files:
# Open all files
print('Testing' , SCREAMING_SNAKE_CASE__ )
if only_modules:
snake_case: Optional[int] = file.split('.' )[0]
try:
snake_case: Dict = getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: str = doctest.DocTestSuite(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = unittest.TextTestRunner().run(SCREAMING_SNAKE_CASE__ )
self.assertIs(len(result.failures ) , 0 )
except AttributeError:
logger.info(F"""{module_identifier} is not a module.""" )
else:
snake_case: List[str] = doctest.testfile(str('..' / directory / file ) , optionflags=doctest.ELLIPSIS )
self.assertIs(result.failed , 0 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = Path('src/transformers' )
snake_case: Optional[int] = 'modeling'
snake_case: List[Any] = [
'modeling_ctrl.py',
'modeling_tf_ctrl.py',
]
self.analyze_directory(SCREAMING_SNAKE_CASE__ , identifier=SCREAMING_SNAKE_CASE__ , ignore_files=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = Path('src/transformers' )
snake_case: str = 'tokenization'
self.analyze_directory(SCREAMING_SNAKE_CASE__ , identifier=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = Path('src/transformers' )
snake_case: Any = 'configuration'
self.analyze_directory(SCREAMING_SNAKE_CASE__ , identifier=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = Path('src/transformers' )
snake_case: str = ['configuration', 'modeling', 'tokenization']
self.analyze_directory(SCREAMING_SNAKE_CASE__ , n_identifier=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = Path('docs/source' )
snake_case: Any = ['favicon.ico']
self.analyze_directory(SCREAMING_SNAKE_CASE__ , ignore_files=SCREAMING_SNAKE_CASE__ , only_modules=SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
import argparse
from typing import Dict
import tensorflow as tf
import torch
from tqdm import tqdm
from transformers import BigBirdPegasusConfig, BigBirdPegasusForConditionalGeneration
__UpperCAmelCase = [
# tf -> hf
("/", "."),
("layer_", "layers."),
("kernel", "weight"),
("beta", "bias"),
("gamma", "weight"),
("pegasus", "model"),
]
__UpperCAmelCase = [
(".output.dense", ".fc2"),
("intermediate.LayerNorm", "final_layer_norm"),
("intermediate.dense", "fc1"),
]
__UpperCAmelCase = (
INIT_COMMON
+ [
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.out_proj"),
("attention.self", "self_attn"),
("attention.encdec.LayerNorm", "encoder_attn_layer_norm"),
("attention.encdec_output.dense", "encoder_attn.out_proj"),
("attention.encdec", "encoder_attn"),
("key", "k_proj"),
("value", "v_proj"),
("query", "q_proj"),
("decoder.LayerNorm", "decoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = (
INIT_COMMON
+ [
("embeddings.word_embeddings", "shared.weight"),
("embeddings.position_embeddings", "embed_positions.weight"),
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.output"),
("attention.self", "self_attn.self"),
("encoder.LayerNorm", "encoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = [
"encdec/key/bias",
"encdec/query/bias",
"encdec/value/bias",
"self/key/bias",
"self/query/bias",
"self/value/bias",
"encdec_output/dense/bias",
"attention/output/dense/bias",
]
def lowerCAmelCase_ ( __A : Dict , __A : List[Any] ):
'''simple docstring'''
for tf_name, hf_name in patterns:
snake_case: List[Any] = k.replace(__A , __A )
return k
def lowerCAmelCase_ ( __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[int] = BigBirdPegasusConfig(**__A )
snake_case: List[Any] = BigBirdPegasusForConditionalGeneration(__A )
snake_case: Any = torch_model.state_dict()
snake_case: Any = {}
# separating decoder weights
snake_case: Optional[Any] = {k: tf_weights[k] for k in tf_weights if k.startswith('pegasus/decoder' )}
snake_case: Any = {k: tf_weights[k] for k in tf_weights if not k.startswith('pegasus/decoder' )}
for k, v in tqdm(decoder_weights.items() , 'tf -> hf conversion' ):
snake_case: List[str] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Any = DECODER_PATTERNS
snake_case: int = rename_state_dict_key(__A , __A )
if new_k not in state_dict:
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: Optional[Any] = v.T
snake_case: Any = torch.from_numpy(__A )
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
for k, v in tqdm(remaining_weights.items() , 'tf -> hf conversion' ):
snake_case: List[Any] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Union[str, Any] = REMAINING_PATTERNS
snake_case: str = rename_state_dict_key(__A , __A )
if new_k not in state_dict and k != "pegasus/embeddings/position_embeddings":
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: int = v.T
snake_case: Any = torch.from_numpy(__A )
if k != "pegasus/embeddings/position_embeddings":
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
snake_case: str = mapping['model.embed_positions.weight']
snake_case: Any = mapping.pop('model.embed_positions.weight' )
snake_case , snake_case: Union[str, Any] = torch_model.load_state_dict(__A , strict=__A )
snake_case: Optional[int] = [
k
for k in missing
if k
not in [
'final_logits_bias',
'model.encoder.embed_tokens.weight',
'model.decoder.embed_tokens.weight',
'lm_head.weight',
]
]
assert unexpected_missing == [], f"""no matches found for the following torch keys {unexpected_missing}"""
assert extra == [], f"""no matches found for the following tf keys {extra}"""
return torch_model
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
snake_case: Tuple = tf.train.list_variables(__A )
snake_case: str = {}
snake_case: List[str] = ['global_step']
for name, shape in tqdm(__A , desc='converting tf checkpoint to dict' ):
snake_case: str = any(pat in name for pat in ignore_name )
if skip_key:
continue
snake_case: Any = tf.train.load_variable(__A , __A )
snake_case: Optional[int] = array
return tf_weights
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict ):
'''simple docstring'''
snake_case: int = get_tf_weights_as_numpy(__A )
snake_case: int = convert_bigbird_pegasus(__A , __A )
torch_model.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--tf_ckpt_path", type=str, help="passed to tf.train.list_variables")
parser.add_argument("--save_dir", default=None, type=str, help="Path to the output PyTorch model.")
__UpperCAmelCase = parser.parse_args()
__UpperCAmelCase = {}
convert_bigbird_pegasus_ckpt_to_pytorch(args.tf_ckpt_path, args.save_dir, config_update=config_update) | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = [
[0, 16, 13, 0, 0, 0],
[0, 0, 10, 12, 0, 0],
[0, 4, 0, 0, 14, 0],
[0, 0, 9, 0, 0, 20],
[0, 0, 0, 7, 0, 4],
[0, 0, 0, 0, 0, 0],
]
def lowerCAmelCase_ ( __A : str , __A : Dict , __A : Dict , __A : Any ):
'''simple docstring'''
snake_case: str = [False] * len(__A )
snake_case: Any = [s]
snake_case: int = True
while queue:
snake_case: Optional[Any] = queue.pop(0 )
for ind in range(len(graph[u] ) ):
if visited[ind] is False and graph[u][ind] > 0:
queue.append(__A )
snake_case: List[str] = True
snake_case: str = u
return visited[t]
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : Dict ):
'''simple docstring'''
snake_case: List[Any] = [-1] * (len(__A ))
snake_case: Any = 0
snake_case: Tuple = []
snake_case: Any = [i[:] for i in graph] # Record original cut, copy.
while bfs(__A , __A , __A , __A ):
snake_case: int = float('Inf' )
snake_case: Any = sink
while s != source:
# Find the minimum value in select path
snake_case: str = min(__A , graph[parent[s]][s] )
snake_case: Dict = parent[s]
max_flow += path_flow
snake_case: Optional[Any] = sink
while v != source:
snake_case: Dict = parent[v]
graph[u][v] -= path_flow
graph[v][u] += path_flow
snake_case: Optional[Any] = parent[v]
for i in range(len(__A ) ):
for j in range(len(graph[0] ) ):
if graph[i][j] == 0 and temp[i][j] > 0:
res.append((i, j) )
return res
if __name__ == "__main__":
print(mincut(test_graph, source=0, sink=5)) | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: str = [0] * len(__A )
snake_case: Tuple = []
snake_case: Tuple = [1] * len(__A )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(__A ) ):
if indegree[i] == 0:
queue.append(__A )
while queue:
snake_case: int = queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
snake_case: Any = long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(__A )
print(max(__A ) )
# Adjacency list of Graph
__UpperCAmelCase = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph) | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 10**9 ):
'''simple docstring'''
snake_case: int = 1
snake_case: List[Any] = 2
snake_case: str = 0
snake_case: Optional[Any] = 0
snake_case: List[Any] = 0
while perimeter <= max_perimeter:
perimeters_sum += perimeter
prev_value += 2 * value
value += prev_value
snake_case: Tuple = 2 * value + 2 if i % 2 == 0 else 2 * value - 2
i += 1
return perimeters_sum
if __name__ == "__main__":
print(F'{solution() = }') | 692 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
import numpy as np
import pytest
from transformers import BertTokenizer, BertTokenizerFast
from transformers.models.bert.tokenization_bert import VOCAB_FILES_NAMES
from transformers.testing_utils import require_vision
from transformers.utils import FEATURE_EXTRACTOR_NAME, is_vision_available
if is_vision_available():
from PIL import Image
from transformers import ChineseCLIPImageProcessor, ChineseCLIPProcessor
@require_vision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = tempfile.mkdtemp()
snake_case: Optional[Any] = [
'[UNK]',
'[CLS]',
'[SEP]',
'[PAD]',
'[MASK]',
'的',
'价',
'格',
'是',
'15',
'便',
'alex',
'##andra',
'，',
'。',
'-',
't',
'shirt',
]
snake_case: Optional[Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
snake_case: Optional[int] = {
'do_resize': True,
'size': {'height': 2_24, 'width': 2_24},
'do_center_crop': True,
'crop_size': {'height': 18, 'width': 18},
'do_normalize': True,
'image_mean': [0.48_14_54_66, 0.4_57_82_75, 0.40_82_10_73],
'image_std': [0.26_86_29_54, 0.26_13_02_58, 0.27_57_77_11],
'do_convert_rgb': True,
}
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , SCREAMING_SNAKE_CASE__ )
with open(self.image_processor_file , 'w' , encoding='utf-8' ) as fp:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizer.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizerFast.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return ChineseCLIPImageProcessor.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Tuple = [Image.fromarray(np.moveaxis(SCREAMING_SNAKE_CASE__ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizer()
snake_case: Union[str, Any] = self.get_rust_tokenizer()
snake_case: Union[str, Any] = self.get_image_processor()
snake_case: List[str] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_slow.save_pretrained(self.tmpdirname )
snake_case: List[str] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname , use_fast=SCREAMING_SNAKE_CASE__ )
snake_case: Any = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_fast.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor_slow.tokenizer.get_vocab() , tokenizer_slow.get_vocab() )
self.assertEqual(processor_fast.tokenizer.get_vocab() , tokenizer_fast.get_vocab() )
self.assertEqual(tokenizer_slow.get_vocab() , tokenizer_fast.get_vocab() )
self.assertIsInstance(processor_slow.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor_slow.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertEqual(processor_fast.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertIsInstance(processor_slow.image_processor , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=self.get_tokenizer() , image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = self.get_tokenizer(cls_token='(CLS)' , sep_token='(SEP)' )
snake_case: Union[str, Any] = self.get_image_processor(do_normalize=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = ChineseCLIPProcessor.from_pretrained(
self.tmpdirname , cls_token='(CLS)' , sep_token='(SEP)' , do_normalize=SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
self.assertIsInstance(processor.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_image_processor()
snake_case: Tuple = self.get_tokenizer()
snake_case: Optional[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.prepare_image_inputs()
snake_case: List[Any] = image_processor(SCREAMING_SNAKE_CASE__ , return_tensors='np' )
snake_case: Dict = processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='np' )
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_image_processor()
snake_case: Optional[int] = self.get_tokenizer()
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Union[str, Any] = processor(text=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer(SCREAMING_SNAKE_CASE__ )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Tuple = self.prepare_image_inputs()
snake_case: Any = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , ['input_ids', 'token_type_ids', 'attention_mask', 'pixel_values'] )
# test if it raises when no input is passed
with pytest.raises(SCREAMING_SNAKE_CASE__ ):
processor()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.get_image_processor()
snake_case: str = self.get_tokenizer()
snake_case: Union[str, Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = [[1, 4, 5, 8, 1, 0, 8], [3, 4, 3, 1, 1, 8, 9]]
snake_case: int = processor.batch_decode(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = 'Alexandra，T-shirt的价格是15便士。'
snake_case: List[Any] = self.prepare_image_inputs()
snake_case: Dict = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , processor.model_input_names ) | 692 | 1 |
'''simple docstring'''
import argparse
import torch
from transformers import OpenAIGPTConfig, OpenAIGPTModel, load_tf_weights_in_openai_gpt
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
def lowerCAmelCase_ ( __A : Any , __A : Union[str, Any] , __A : List[str] ):
'''simple docstring'''
if openai_config_file == "":
snake_case: str = OpenAIGPTConfig()
else:
snake_case: Tuple = OpenAIGPTConfig.from_json_file(__A )
snake_case: Union[str, Any] = OpenAIGPTModel(__A )
# Load weights from numpy
load_tf_weights_in_openai_gpt(__A , __A , __A )
# Save pytorch-model
snake_case: Optional[int] = pytorch_dump_folder_path + '/' + WEIGHTS_NAME
snake_case: Union[str, Any] = pytorch_dump_folder_path + '/' + CONFIG_NAME
print(f"""Save PyTorch model to {pytorch_weights_dump_path}""" )
torch.save(model.state_dict() , __A )
print(f"""Save configuration file to {pytorch_config_dump_path}""" )
with open(__A , 'w' , encoding='utf-8' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--openai_checkpoint_folder_path",
default=None,
type=str,
required=True,
help="Path to the TensorFlow checkpoint path.",
)
parser.add_argument(
"--pytorch_dump_folder_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
parser.add_argument(
"--openai_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained OpenAI model. \n"
"This specifies the model architecture."
),
)
__UpperCAmelCase = parser.parse_args()
convert_openai_checkpoint_to_pytorch(
args.openai_checkpoint_folder_path, args.openai_config_file, args.pytorch_dump_folder_path
) | 692 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/swinv2-tiny-patch4-window8-256": (
"https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256/resolve/main/config.json"
),
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "swinv2"
__UpperCamelCase = {
"num_attention_heads": "num_heads",
"num_hidden_layers": "num_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=96 , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[3, 6, 12, 24] , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=32 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: int = image_size
snake_case: Union[str, Any] = patch_size
snake_case: List[str] = num_channels
snake_case: Tuple = embed_dim
snake_case: str = depths
snake_case: Any = len(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = num_heads
snake_case: Optional[int] = window_size
snake_case: Any = mlp_ratio
snake_case: Optional[int] = qkv_bias
snake_case: Union[str, Any] = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: Dict = drop_path_rate
snake_case: List[str] = hidden_act
snake_case: int = use_absolute_embeddings
snake_case: Any = layer_norm_eps
snake_case: Dict = initializer_range
snake_case: List[Any] = encoder_stride
# we set the hidden_size attribute in order to make Swinv2 work with VisionEncoderDecoderModel
# this indicates the channel dimension after the last stage of the model
snake_case: Tuple = int(embed_dim * 2 ** (len(SCREAMING_SNAKE_CASE__ ) - 1) )
snake_case: Union[str, Any] = (0, 0, 0, 0) | 692 | 1 |
'''simple docstring'''
import argparse
from typing import List
import evaluate
import numpy as np
import torch
from datasets import DatasetDict, load_dataset
# New Code #
# We'll be using StratifiedKFold for this example
from sklearn.model_selection import StratifiedKFold
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate,
# specifically showcasing how to perform Cross Validation,
# and builds off the `nlp_example.py` script.
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To help focus on the differences in the code, building `DataLoaders`
# was refactored into its own function.
# New additions from the base script can be found quickly by
# looking for the # New Code # tags
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
__UpperCAmelCase = 16
__UpperCAmelCase = 32
def lowerCAmelCase_ ( __A : Accelerator , __A : DatasetDict , __A : List[int] , __A : List[int] , __A : int = 16 ):
'''simple docstring'''
snake_case: int = AutoTokenizer.from_pretrained('bert-base-cased' )
snake_case: Tuple = DatasetDict(
{
'train': dataset['train'].select(__A ),
'validation': dataset['train'].select(__A ),
'test': dataset['validation'],
} )
def tokenize_function(__A : List[Any] ):
# max_length=None => use the model max length (it's actually the default)
snake_case: Any = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__A , max_length=__A )
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
snake_case: Union[str, Any] = datasets.map(
__A , batched=__A , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
snake_case: Optional[Any] = tokenized_datasets.rename_column('label' , 'labels' )
def collate_fn(__A : List[str] ):
# On TPU it's best to pad everything to the same length or training will be very slow.
snake_case: Tuple = 1_28 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
snake_case: Union[str, Any] = 16
elif accelerator.mixed_precision != "no":
snake_case: List[str] = 8
else:
snake_case: List[str] = None
return tokenizer.pad(
__A , padding='longest' , max_length=__A , pad_to_multiple_of=__A , return_tensors='pt' , )
# Instantiate dataloaders.
snake_case: int = DataLoader(
tokenized_datasets['train'] , shuffle=__A , collate_fn=__A , batch_size=__A )
snake_case: Union[str, Any] = DataLoader(
tokenized_datasets['validation'] , shuffle=__A , collate_fn=__A , batch_size=__A )
snake_case: Union[str, Any] = DataLoader(
tokenized_datasets['test'] , shuffle=__A , collate_fn=__A , batch_size=__A )
return train_dataloader, eval_dataloader, test_dataloader
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: List[Any] = []
# Download the dataset
snake_case: Optional[Any] = load_dataset('glue' , 'mrpc' )
# Create our splits
snake_case: Dict = StratifiedKFold(n_splits=int(args.num_folds ) )
# Initialize accelerator
snake_case: Optional[int] = Accelerator(cpu=args.cpu , mixed_precision=args.mixed_precision )
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
snake_case: List[Any] = config['lr']
snake_case: Any = int(config['num_epochs'] )
snake_case: List[str] = int(config['seed'] )
snake_case: int = int(config['batch_size'] )
snake_case: List[Any] = evaluate.load('glue' , 'mrpc' )
# If the batch size is too big we use gradient accumulation
snake_case: int = 1
if batch_size > MAX_GPU_BATCH_SIZE and accelerator.distributed_type != DistributedType.TPU:
snake_case: str = batch_size // MAX_GPU_BATCH_SIZE
snake_case: Optional[int] = MAX_GPU_BATCH_SIZE
set_seed(__A )
# New Code #
# Create our folds:
snake_case: int = kfold.split(np.zeros(datasets['train'].num_rows ) , datasets['train']['label'] )
snake_case: Tuple = []
# Iterate over them
for i, (train_idxs, valid_idxs) in enumerate(__A ):
snake_case , snake_case , snake_case: List[str] = get_fold_dataloaders(
__A , __A , __A , __A , )
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
snake_case: List[str] = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__A )
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
snake_case: Any = model.to(accelerator.device )
# Instantiate optimizer
snake_case: Dict = AdamW(params=model.parameters() , lr=__A )
# Instantiate scheduler
snake_case: Any = get_linear_schedule_with_warmup(
optimizer=__A , num_warmup_steps=1_00 , num_training_steps=(len(__A ) * num_epochs) // gradient_accumulation_steps , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
snake_case , snake_case , snake_case , snake_case , snake_case: Optional[int] = accelerator.prepare(
__A , __A , __A , __A , __A )
# Now we train the model
for epoch in range(__A ):
model.train()
for step, batch in enumerate(__A ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
snake_case: List[Any] = model(**__A )
snake_case: List[str] = outputs.loss
snake_case: List[Any] = loss / gradient_accumulation_steps
accelerator.backward(__A )
if step % gradient_accumulation_steps == 0:
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__A ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
with torch.no_grad():
snake_case: Any = model(**__A )
snake_case: Tuple = outputs.logits.argmax(dim=-1 )
snake_case , snake_case: Optional[int] = accelerator.gather_for_metrics((predictions, batch['labels']) )
metric.add_batch(
predictions=__A , references=__A , )
snake_case: int = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __A )
# New Code #
# We also run predictions on the test set at the very end
snake_case: str = []
for step, batch in enumerate(__A ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
with torch.no_grad():
snake_case: Any = model(**__A )
snake_case: Union[str, Any] = outputs.logits
snake_case , snake_case: Dict = accelerator.gather_for_metrics((predictions, batch['labels']) )
fold_predictions.append(predictions.cpu() )
if i == 0:
# We need all of the test predictions
test_references.append(references.cpu() )
# Use accelerator.print to print only on the main process.
test_predictions.append(torch.cat(__A , dim=0 ) )
# We now need to release all our memory and get rid of the current model, optimizer, etc
accelerator.free_memory()
# New Code #
# Finally we check the accuracy of our folded results:
snake_case: Optional[int] = torch.cat(__A , dim=0 )
snake_case: List[str] = torch.stack(__A , dim=0 ).sum(dim=0 ).div(int(args.num_folds ) ).argmax(dim=-1 )
snake_case: Dict = metric.compute(predictions=__A , references=__A )
accelerator.print('Average test metrics from all folds:' , __A )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Any = argparse.ArgumentParser(description='Simple example of training script.' )
parser.add_argument(
'--mixed_precision' , type=__A , default=__A , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.' )
# New Code #
parser.add_argument('--num_folds' , type=__A , default=3 , help='The number of splits to perform across the dataset' )
snake_case: Optional[Any] = parser.parse_args()
snake_case: Union[str, Any] = {'lr': 2E-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__A , __A )
if __name__ == "__main__":
main() | 692 |
'''simple docstring'''
import os
import sys
import unittest
__UpperCAmelCase = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__))))
sys.path.append(os.path.join(git_repo_path, "utils"))
import check_dummies # noqa: E402
from check_dummies import create_dummy_files, create_dummy_object, find_backend, read_init # noqa: E402
# Align TRANSFORMERS_PATH in check_dummies with the current path
__UpperCAmelCase = os.path.join(git_repo_path, "src", "transformers")
__UpperCAmelCase = "\n{0} = None\n"
__UpperCAmelCase = "\nclass {0}(metaclass=DummyObject):\n _backends = {1}\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, {1})\n"
__UpperCAmelCase = "\ndef {0}(*args, **kwargs):\n requires_backends({0}, {1})\n"
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = find_backend(' _import_structure["models.albert"].append("AlbertTokenizerFast")' )
self.assertIsNone(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = find_backend(' if not is_tokenizers_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tokenizers' )
snake_case: List[Any] = find_backend(' if not is_tensorflow_text_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tensorflow_text' )
snake_case: int = find_backend(' if not (is_sentencepiece_available() and is_tokenizers_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers' )
snake_case: Optional[Any] = find_backend(
' if not (is_sentencepiece_available() and is_tensorflow_text_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tensorflow_text' )
snake_case: Dict = find_backend(
' if not (is_sentencepiece_available() and is_tokenizers_available() and is_vision_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers_and_vision' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = read_init()
# We don't assert on the exact list of keys to allow for smooth grow of backend-specific objects
self.assertIn('torch' , SCREAMING_SNAKE_CASE__ )
self.assertIn('tensorflow_text' , SCREAMING_SNAKE_CASE__ )
self.assertIn('sentencepiece_and_tokenizers' , SCREAMING_SNAKE_CASE__ )
# Likewise, we can't assert on the exact content of a key
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertModel' , objects['tf'] )
self.assertIn('FlaxBertModel' , objects['flax'] )
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertTokenizer' , objects['tensorflow_text'] )
self.assertIn('convert_slow_tokenizer' , objects['sentencepiece_and_tokenizers'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = create_dummy_object('CONSTANT' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , '\nCONSTANT = None\n' )
snake_case: Any = create_dummy_object('function' , '\'torch\'' )
self.assertEqual(
SCREAMING_SNAKE_CASE__ , '\ndef function(*args, **kwargs):\n requires_backends(function, \'torch\')\n' )
snake_case: Optional[int] = '\nclass FakeClass(metaclass=DummyObject):\n _backends = \'torch\'\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, \'torch\')\n'
snake_case: Tuple = create_dummy_object('FakeClass' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = '# This file is autogenerated by the command `make fix-copies`, do not edit.\nfrom ..utils import DummyObject, requires_backends\n\n\nCONSTANT = None\n\n\ndef function(*args, **kwargs):\n requires_backends(function, ["torch"])\n\n\nclass FakeClass(metaclass=DummyObject):\n _backends = ["torch"]\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, ["torch"])\n'
snake_case: Optional[int] = create_dummy_files({'torch': ['CONSTANT', 'function', 'FakeClass']} )
self.assertEqual(dummy_files['torch'] , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
import unittest
from diffusers import FlaxAutoencoderKL
from diffusers.utils import is_flax_available
from diffusers.utils.testing_utils import require_flax
from .test_modeling_common_flax import FlaxModelTesterMixin
if is_flax_available():
import jax
@require_flax
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = FlaxAutoencoderKL
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = 4
snake_case: Optional[int] = 3
snake_case: List[Any] = (32, 32)
snake_case: List[Any] = jax.random.PRNGKey(0 )
snake_case: Dict = jax.random.uniform(SCREAMING_SNAKE_CASE__ , ((batch_size, num_channels) + sizes) )
return {"sample": image, "prng_key": prng_key}
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = {
'block_out_channels': [32, 64],
'in_channels': 3,
'out_channels': 3,
'down_block_types': ['DownEncoderBlock2D', 'DownEncoderBlock2D'],
'up_block_types': ['UpDecoderBlock2D', 'UpDecoderBlock2D'],
'latent_channels': 4,
}
snake_case: Any = self.dummy_input
return init_dict, inputs_dict | 692 |
'''simple docstring'''
import os
import warnings
from typing import List, Optional
from ...tokenization_utils_base import BatchEncoding
from ...utils import logging
from .configuration_rag import RagConfig
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = question_encoder
snake_case: Union[str, Any] = generator
snake_case: Optional[int] = self.question_encoder
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if os.path.isfile(SCREAMING_SNAKE_CASE__ ):
raise ValueError(F"""Provided path ({save_directory}) should be a directory, not a file""" )
os.makedirs(SCREAMING_SNAKE_CASE__ , exist_ok=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'question_encoder_tokenizer' )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'generator_tokenizer' )
self.question_encoder.save_pretrained(SCREAMING_SNAKE_CASE__ )
self.generator.save_pretrained(SCREAMING_SNAKE_CASE__ )
@classmethod
def _UpperCamelCase ( cls , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
from ..auto.tokenization_auto import AutoTokenizer
snake_case: int = kwargs.pop('config' , SCREAMING_SNAKE_CASE__ )
if config is None:
snake_case: str = RagConfig.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.question_encoder , subfolder='question_encoder_tokenizer' )
snake_case: Dict = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.generator , subfolder='generator_tokenizer' )
return cls(question_encoder=SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ )
def __call__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.current_tokenizer(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.batch_decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.question_encoder
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.generator
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "longest" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
warnings.warn(
'`prepare_seq2seq_batch` is deprecated and will be removed in version 5 of 🤗 Transformers. Use the '
'regular `__call__` method to prepare your inputs and the tokenizer under the `with_target_tokenizer` '
'context manager to prepare your targets. See the documentation of your specific tokenizer for more '
'details' , SCREAMING_SNAKE_CASE__ , )
if max_length is None:
snake_case: Optional[Any] = self.current_tokenizer.model_max_length
snake_case: int = self(
SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if tgt_texts is None:
return model_inputs
# Process tgt_texts
if max_target_length is None:
snake_case: Any = self.current_tokenizer.model_max_length
snake_case: List[str] = self(
text_target=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: Dict = labels['input_ids']
return model_inputs | 692 | 1 |
'''simple docstring'''
__UpperCAmelCase = {
"joule": 1.0,
"kilojoule": 1_000,
"megajoule": 1_000_000,
"gigajoule": 1_000_000_000,
"wattsecond": 1.0,
"watthour": 3_600,
"kilowatthour": 3_600_000,
"newtonmeter": 1.0,
"calorie_nutr": 4_186.8,
"kilocalorie_nutr": 4_186_800.00,
"electronvolt": 1.6_02_17_66_34E-19,
"britishthermalunit_it": 1_055.05_585,
"footpound": 1.355818,
}
def lowerCAmelCase_ ( __A : str , __A : str , __A : float ):
'''simple docstring'''
if to_type not in ENERGY_CONVERSION or from_type not in ENERGY_CONVERSION:
snake_case: List[str] = (
f"""Incorrect 'from_type' or 'to_type' value: {from_type!r}, {to_type!r}\n"""
f"""Valid values are: {', '.join(__A )}"""
)
raise ValueError(__A )
return value * ENERGY_CONVERSION[from_type] / ENERGY_CONVERSION[to_type]
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 |
'''simple docstring'''
import importlib
import os
import fsspec
import pytest
from fsspec import register_implementation
from fsspec.registry import _registry as _fsspec_registry
from datasets.filesystems import COMPRESSION_FILESYSTEMS, HfFileSystem, extract_path_from_uri, is_remote_filesystem
from .utils import require_lza, require_zstandard
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
assert "mock" in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert "mock" not in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'mock-s3-bucket'
snake_case: int = f"""s3://{mock_bucket}"""
snake_case: Any = extract_path_from_uri(__A )
assert dataset_path.startswith('s3://' ) is False
snake_case: Union[str, Any] = './local/path'
snake_case: Union[str, Any] = extract_path_from_uri(__A )
assert dataset_path == new_dataset_path
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
snake_case: List[str] = is_remote_filesystem(__A )
assert is_remote is True
snake_case: int = fsspec.filesystem('file' )
snake_case: int = is_remote_filesystem(__A )
assert is_remote is False
@pytest.mark.parametrize('compression_fs_class' , __A )
def lowerCAmelCase_ ( __A : Optional[int] , __A : int , __A : str , __A : Optional[Any] , __A : List[str] , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: Optional[Any] = {'gzip': gz_file, 'xz': xz_file, 'zstd': zstd_file, 'bz2': bza_file, 'lz4': lza_file}
snake_case: Optional[int] = input_paths[compression_fs_class.protocol]
if input_path is None:
snake_case: str = f"""for '{compression_fs_class.protocol}' compression protocol, """
if compression_fs_class.protocol == "lz4":
reason += require_lza.kwargs["reason"]
elif compression_fs_class.protocol == "zstd":
reason += require_zstandard.kwargs["reason"]
pytest.skip(__A )
snake_case: List[str] = fsspec.filesystem(compression_fs_class.protocol , fo=__A )
assert isinstance(__A , __A )
snake_case: Any = os.path.basename(__A )
snake_case: int = expected_filename[: expected_filename.rindex('.' )]
assert fs.glob('*' ) == [expected_filename]
with fs.open(__A , 'r' , encoding='utf-8' ) as f, open(__A , encoding='utf-8' ) as expected_file:
assert f.read() == expected_file.read()
@pytest.mark.parametrize('protocol' , ['zip', 'gzip'] )
def lowerCAmelCase_ ( __A : Any , __A : int , __A : int ):
'''simple docstring'''
snake_case: List[str] = {'zip': zip_jsonl_path, 'gzip': jsonl_gz_path}
snake_case: str = compressed_file_paths[protocol]
snake_case: Dict = 'dataset.jsonl'
snake_case: Optional[Any] = f"""{protocol}://{member_file_path}::{compressed_file_path}"""
snake_case , *snake_case: List[Any] = fsspec.get_fs_token_paths(__A )
assert fs.isfile(__A )
assert not fs.isfile('non_existing_' + member_file_path )
@pytest.mark.integration
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Union[str, Any] , __A : List[Any] ):
'''simple docstring'''
snake_case: Tuple = hf_api.dataset_info(__A , token=__A )
snake_case: List[str] = HfFileSystem(repo_info=__A , token=__A )
assert sorted(hffs.glob('*' ) ) == [".gitattributes", "data"]
assert hffs.isdir('data' )
assert hffs.isfile('.gitattributes' ) and hffs.isfile('data/text_data.txt' )
with open(__A ) as f:
assert hffs.open('data/text_data.txt' , 'r' ).read() == f.read()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'bz2'
# Import module
import datasets.filesystems
# Overwrite protocol and reload
register_implementation(__A , __A , clobber=__A )
with pytest.warns(__A ) as warning_info:
importlib.reload(datasets.filesystems )
assert len(__A ) == 1
assert (
str(warning_info[0].message )
== f"""A filesystem protocol was already set for {protocol} and will be overwritten."""
) | 692 | 1 |
'''simple docstring'''
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import TYPE_CHECKING
# rely on isort to merge the imports
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
__UpperCAmelCase = {
"configuration_vivit": ["VIVIT_PRETRAINED_CONFIG_ARCHIVE_MAP", "VivitConfig"],
}
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["VivitImageProcessor"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"VIVIT_PRETRAINED_MODEL_ARCHIVE_LIST",
"VivitModel",
"VivitPreTrainedModel",
"VivitForVideoClassification",
]
if TYPE_CHECKING:
from .configuration_vivit import VIVIT_PRETRAINED_CONFIG_ARCHIVE_MAP, VivitConfig
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .image_processing_vivit import VivitImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_vivit import (
VIVIT_PRETRAINED_MODEL_ARCHIVE_LIST,
VivitForVideoClassification,
VivitModel,
VivitPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
import logging
import os
import random
import sys
from dataclasses import dataclass, field
from typing import Optional
import datasets
import numpy as np
import pandas as pd
from datasets import load_dataset
import transformers
from transformers import (
AutoConfig,
BartForSequenceClassification,
DataCollatorWithPadding,
EvalPrediction,
HfArgumentParser,
TapexTokenizer,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.17.0.dev0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
__UpperCAmelCase = logging.getLogger(__name__)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The name of the dataset to use (via the datasets library)."} )
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} , )
__UpperCamelCase = field(
default=1024 , metadata={
"help": (
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Overwrite the cached preprocessed datasets or not."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Whether to pad all samples to `max_seq_length`. "
"If False, will pad the samples dynamically when batching to the maximum length in the batch."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of prediction examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the training data."} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the validation data."} )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A csv or a json file containing the test data."} )
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.dataset_name is not None:
pass
elif self.train_file is None or self.validation_file is None:
raise ValueError('Need either a GLUE task, a training/validation file or a dataset name.' )
else:
snake_case: str = self.train_file.split('.' )[-1]
assert train_extension in ["csv", "json"], "`train_file` should be a csv or a json file."
snake_case: Optional[Any] = self.validation_file.split('.' )[-1]
assert (
validation_extension == train_extension
), "`validation_file` should have the same extension (csv or json) as `train_file`."
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."} , )
__UpperCamelCase = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case , snake_case , snake_case: Union[str, Any] = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case , snake_case , snake_case: str = parser.parse_args_into_dataclasses()
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
snake_case: Tuple = training_args.get_process_log_level()
logger.setLevel(__A )
datasets.utils.logging.set_verbosity(__A )
transformers.utils.logging.set_verbosity(__A )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"""
+ f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" )
logger.info(f"""Training/evaluation parameters {training_args}""" )
# Detecting last checkpoint.
snake_case: Any = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case: List[Any] = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"""Output directory ({training_args.output_dir}) already exists and is not empty. """
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Set seed before initializing model.
set_seed(training_args.seed )
# Get the datasets: you can either provide your own CSV/JSON training and evaluation files (see below)
# or specify a GLUE benchmark task (the dataset will be downloaded automatically from the datasets Hub).
#
# For JSON files, this script will use the `question` column for the input question and `table` column for the corresponding table.
#
# If the CSVs/JSONs contain only one non-label column, the script does single sentence classification on this
# single column. You can easily tweak this behavior (see below)
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
snake_case: int = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from your local files.
# CSV/JSON training and evaluation files are needed.
snake_case: Optional[int] = {'train': data_args.train_file, 'validation': data_args.validation_file}
# Get the test dataset: you can provide your own CSV/JSON test file (see below)
# when you use `do_predict` without specifying a GLUE benchmark task.
if training_args.do_predict:
if data_args.test_file is not None:
snake_case: Tuple = data_args.train_file.split('.' )[-1]
snake_case: Union[str, Any] = data_args.test_file.split('.' )[-1]
assert (
test_extension == train_extension
), "`test_file` should have the same extension (csv or json) as `train_file`."
snake_case: Union[str, Any] = data_args.test_file
else:
raise ValueError('Need either a GLUE task or a test file for `do_predict`.' )
for key in data_files.keys():
logger.info(f"""load a local file for {key}: {data_files[key]}""" )
if data_args.train_file.endswith('.csv' ):
# Loading a dataset from local csv files
snake_case: List[Any] = load_dataset('csv' , data_files=__A , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from local json files
snake_case: Optional[Any] = load_dataset('json' , data_files=__A , cache_dir=model_args.cache_dir )
# See more about loading any type of standard or custom dataset at
# https://huggingface.co/docs/datasets/loading_datasets.html.
# Labels
snake_case: Tuple = raw_datasets['train'].features['label'].names
snake_case: List[str] = len(__A )
# Load pretrained model and tokenizer
#
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case: Tuple = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# load tapex tokenizer
snake_case: List[str] = TapexTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , add_prefix_space=__A , )
snake_case: Union[str, Any] = BartForSequenceClassification.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# Padding strategy
if data_args.pad_to_max_length:
snake_case: int = 'max_length'
else:
# We will pad later, dynamically at batch creation, to the max sequence length in each batch
snake_case: Union[str, Any] = False
# Some models have set the order of the labels to use, so let's make sure we do use it.
snake_case: Optional[Any] = {'Refused': 0, 'Entailed': 1}
snake_case: List[Any] = {0: 'Refused', 1: 'Entailed'}
if data_args.max_seq_length > tokenizer.model_max_length:
logger.warning(
f"""The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"""
f"""model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}.""" )
snake_case: List[str] = min(data_args.max_seq_length , tokenizer.model_max_length )
def preprocess_tabfact_function(__A : Any ):
# Tokenize the texts
def _convert_table_text_to_pandas(__A : Dict ):
snake_case: str = [_table_row.split('#' ) for _table_row in _table_text.strip('\n' ).split('\n' )]
snake_case: List[str] = pd.DataFrame.from_records(_table_content[1:] , columns=_table_content[0] )
return _table_pd
snake_case: str = examples['statement']
snake_case: int = list(map(_convert_table_text_to_pandas , examples['table_text'] ) )
snake_case: List[Any] = tokenizer(__A , __A , padding=__A , max_length=__A , truncation=__A )
snake_case: List[Any] = examples['label']
return result
with training_args.main_process_first(desc='dataset map pre-processing' ):
snake_case: int = raw_datasets.map(
__A , batched=__A , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on dataset' , )
if training_args.do_train:
if "train" not in raw_datasets:
raise ValueError('--do_train requires a train dataset' )
snake_case: List[str] = raw_datasets['train']
if data_args.max_train_samples is not None:
snake_case: Tuple = train_dataset.select(range(data_args.max_train_samples ) )
if training_args.do_eval:
if "validation" not in raw_datasets and "validation_matched" not in raw_datasets:
raise ValueError('--do_eval requires a validation dataset' )
snake_case: Any = raw_datasets['validation']
if data_args.max_eval_samples is not None:
snake_case: Optional[int] = eval_dataset.select(range(data_args.max_eval_samples ) )
if training_args.do_predict or data_args.test_file is not None:
if "test" not in raw_datasets and "test_matched" not in raw_datasets:
raise ValueError('--do_predict requires a test dataset' )
snake_case: str = raw_datasets['test']
if data_args.max_predict_samples is not None:
snake_case: List[str] = predict_dataset.select(range(data_args.max_predict_samples ) )
# Log a few random samples from the training set:
if training_args.do_train:
for index in random.sample(range(len(__A ) ) , 3 ):
logger.info(f"""Sample {index} of the training set: {train_dataset[index]}.""" )
# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def compute_metrics(__A : EvalPrediction ):
snake_case: int = p.predictions[0] if isinstance(p.predictions , __A ) else p.predictions
snake_case: List[str] = np.argmax(__A , axis=1 )
return {"accuracy": (preds == p.label_ids).astype(np.floataa ).mean().item()}
# Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
if data_args.pad_to_max_length:
snake_case: str = default_data_collator
elif training_args.fpaa:
snake_case: List[str] = DataCollatorWithPadding(__A , pad_to_multiple_of=8 )
else:
snake_case: List[Any] = None
# Initialize our Trainer
snake_case: List[str] = Trainer(
model=__A , args=__A , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , compute_metrics=__A , tokenizer=__A , data_collator=__A , )
# Training
if training_args.do_train:
snake_case: Optional[int] = None
if training_args.resume_from_checkpoint is not None:
snake_case: str = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
snake_case: Optional[Any] = last_checkpoint
snake_case: Union[str, Any] = trainer.train(resume_from_checkpoint=__A )
snake_case: List[Any] = train_result.metrics
snake_case: List[Any] = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(__A )
)
snake_case: Optional[Any] = min(__A , len(__A ) )
trainer.save_model() # Saves the tokenizer too for easy upload
trainer.log_metrics('train' , __A )
trainer.save_metrics('train' , __A )
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info('*** Evaluate ***' )
snake_case: Dict = trainer.evaluate(eval_dataset=__A )
snake_case: Optional[int] = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(__A )
snake_case: Dict = min(__A , len(__A ) )
trainer.log_metrics('eval' , __A )
trainer.save_metrics('eval' , __A )
if training_args.do_predict:
logger.info('*** Predict ***' )
# Removing the `label` columns because it contains -1 and Trainer won't like that.
snake_case: Optional[int] = predict_dataset.remove_columns('label' )
snake_case: str = trainer.predict(__A , metric_key_prefix='predict' ).predictions
snake_case: Any = np.argmax(__A , axis=1 )
snake_case: int = os.path.join(training_args.output_dir , 'predict_results_tabfact.txt' )
if trainer.is_world_process_zero():
with open(__A , 'w' ) as writer:
logger.info('***** Predict Results *****' )
writer.write('index\tprediction\n' )
for index, item in enumerate(__A ):
snake_case: int = label_list[item]
writer.write(f"""{index}\t{item}\n""" )
snake_case: Optional[int] = {'finetuned_from': model_args.model_name_or_path, 'tasks': 'text-classification'}
if training_args.push_to_hub:
trainer.push_to_hub(**__A )
else:
trainer.create_model_card(**__A )
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
main()
if __name__ == "__main__":
main() | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
import unittest
from transformers import is_tf_available
from transformers.testing_utils import require_tf, slow
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import numpy
import tensorflow as tf
from transformers import (
TF_DPR_CONTEXT_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST,
TF_DPR_QUESTION_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST,
TF_DPR_READER_PRETRAINED_MODEL_ARCHIVE_LIST,
BertConfig,
DPRConfig,
TFDPRContextEncoder,
TFDPRQuestionEncoder,
TFDPRReader,
)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=13 , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=99 , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=37 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=0 , ):
'''simple docstring'''
snake_case: int = parent
snake_case: List[str] = batch_size
snake_case: Optional[Any] = seq_length
snake_case: Any = is_training
snake_case: Optional[Any] = use_input_mask
snake_case: List[str] = use_token_type_ids
snake_case: Any = use_labels
snake_case: Dict = vocab_size
snake_case: str = hidden_size
snake_case: str = num_hidden_layers
snake_case: Union[str, Any] = num_attention_heads
snake_case: Optional[Any] = intermediate_size
snake_case: Union[str, Any] = hidden_act
snake_case: List[Any] = hidden_dropout_prob
snake_case: Any = attention_probs_dropout_prob
snake_case: List[Any] = max_position_embeddings
snake_case: Union[str, Any] = type_vocab_size
snake_case: Any = type_sequence_label_size
snake_case: Dict = initializer_range
snake_case: List[str] = num_labels
snake_case: Tuple = num_choices
snake_case: Union[str, Any] = scope
snake_case: Dict = projection_dim
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case: Optional[Any] = None
if self.use_input_mask:
# follow test_modeling_tf_ctrl.py
snake_case: List[Any] = random_attention_mask([self.batch_size, self.seq_length] )
snake_case: Tuple = None
if self.use_token_type_ids:
snake_case: List[Any] = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case: int = None
snake_case: Optional[Any] = None
snake_case: List[str] = None
if self.use_labels:
snake_case: int = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case: str = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
snake_case: Optional[Any] = ids_tensor([self.batch_size] , self.num_choices )
snake_case: List[str] = BertConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=SCREAMING_SNAKE_CASE__ , initializer_range=self.initializer_range , )
snake_case: List[str] = DPRConfig(projection_dim=self.projection_dim , **config.to_dict() )
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = TFDPRContextEncoder(config=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: str = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.pooler_output.shape , (self.batch_size, self.projection_dim or self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Union[str, Any] = TFDPRQuestionEncoder(config=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: int = model(SCREAMING_SNAKE_CASE__ , token_type_ids=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = model(SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.pooler_output.shape , (self.batch_size, self.projection_dim or self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = TFDPRReader(config=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.relevance_logits.shape , (self.batch_size,) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = self.prepare_config_and_inputs()
(
(
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) , (
snake_case
) ,
): List[str] = config_and_inputs
snake_case: Dict = {'input_ids': input_ids}
return config, inputs_dict
@require_tf
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (
(
TFDPRContextEncoder,
TFDPRQuestionEncoder,
TFDPRReader,
)
if is_tf_available()
else ()
)
__UpperCamelCase = {"feature-extraction": TFDPRQuestionEncoder} if is_tf_available() else {}
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = TFDPRModelTester(self )
snake_case: Tuple = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , hidden_size=37 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_dpr_context_encoder(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_dpr_question_encoder(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_dpr_reader(*SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
for model_name in TF_DPR_CONTEXT_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: Optional[int] = TFDPRContextEncoder.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
for model_name in TF_DPR_CONTEXT_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: int = TFDPRContextEncoder.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
for model_name in TF_DPR_QUESTION_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: Tuple = TFDPRQuestionEncoder.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
for model_name in TF_DPR_READER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
snake_case: List[str] = TFDPRReader.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
@require_tf
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = TFDPRQuestionEncoder.from_pretrained('facebook/dpr-question_encoder-single-nq-base' )
snake_case: Dict = tf.constant(
[[1_01, 75_92, 10_10, 20_03, 20_26, 38_99, 1_01_40, 10_29, 1_02]] ) # [CLS] hello, is my dog cute? [SEP]
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ )[0] # embedding shape = (1, 768)
# compare the actual values for a slice.
snake_case: str = tf.constant(
[
[
0.03_23_62_53,
0.12_75_33_35,
0.16_81_85_09,
0.00_27_97_86,
0.3_89_69_33,
0.24_26_49_45,
0.2_17_89_71,
-0.02_33_52_27,
-0.08_48_19_59,
-0.14_32_41_17,
]
] )
self.assertTrue(numpy.allclose(output[:, :10].numpy() , expected_slice.numpy() , atol=1E-4 ) ) | 692 |
'''simple docstring'''
import math
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or number % 2 == 0 or number % 3 == 0:
# Negatives, 0, 1, all even numbers, all multiples of 3 are not primes
return False
# All primes number are in format of 6k +/- 1
for i in range(5 , int(math.sqrt(__A ) + 1 ) , 6 ):
if number % i == 0 or number % (i + 2) == 0:
return False
return True
def lowerCAmelCase_ ( __A : float = 0.1 ):
'''simple docstring'''
snake_case: Optional[int] = 3
snake_case: int = 3
while primes / (2 * j - 1) >= ratio:
for i in range(j * j + j + 1 , (j + 2) * (j + 2) , j + 1 ):
primes += is_prime(__A )
j += 2
return j
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
from __future__ import annotations
import copy
import inspect
import json
import math
import os
import tempfile
import unittest
from importlib import import_module
import numpy as np
from transformers import ViTMAEConfig
from transformers.file_utils import cached_property, is_tf_available, is_vision_available
from transformers.testing_utils import require_tf, require_vision, slow
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFViTMAEForPreTraining, TFViTMAEModel
if is_vision_available():
from PIL import Image
from transformers import ViTImageProcessor
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=13 , SCREAMING_SNAKE_CASE__=30 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=2 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=37 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=10 , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=0.6 , SCREAMING_SNAKE_CASE__=None , ):
'''simple docstring'''
snake_case: Optional[int] = parent
snake_case: List[Any] = batch_size
snake_case: int = image_size
snake_case: Optional[Any] = patch_size
snake_case: str = num_channels
snake_case: Optional[int] = is_training
snake_case: Dict = use_labels
snake_case: List[Any] = hidden_size
snake_case: Optional[Any] = num_hidden_layers
snake_case: Tuple = num_attention_heads
snake_case: List[Any] = intermediate_size
snake_case: List[str] = hidden_act
snake_case: str = hidden_dropout_prob
snake_case: Union[str, Any] = attention_probs_dropout_prob
snake_case: Dict = type_sequence_label_size
snake_case: str = initializer_range
snake_case: Union[str, Any] = mask_ratio
snake_case: Dict = scope
# in ViTMAE, the expected sequence length = (num_patches + 1) * (1 - config.mask_ratio), rounded above
# (we add 1 for the [CLS] token)
snake_case: List[Any] = (image_size // patch_size) ** 2
snake_case: Optional[int] = int(math.ceil((1 - mask_ratio) * (num_patches + 1) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
snake_case: List[Any] = None
if self.use_labels:
snake_case: Dict = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case: List[Any] = self.get_config()
return config, pixel_values, labels
def _UpperCamelCase ( self ):
'''simple docstring'''
return ViTMAEConfig(
image_size=self.image_size , patch_size=self.patch_size , num_channels=self.num_channels , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , decoder_hidden_size=self.hidden_size , decoder_num_hidden_layers=self.num_hidden_layers , decoder_num_attention_heads=self.num_attention_heads , decoder_intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , is_decoder=SCREAMING_SNAKE_CASE__ , initializer_range=self.initializer_range , mask_ratio=self.mask_ratio , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = TFViTMAEModel(config=SCREAMING_SNAKE_CASE__ )
snake_case: Any = model(SCREAMING_SNAKE_CASE__ , training=SCREAMING_SNAKE_CASE__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = TFViTMAEForPreTraining(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = model(SCREAMING_SNAKE_CASE__ , training=SCREAMING_SNAKE_CASE__ )
# expected sequence length = num_patches
snake_case: Any = (self.image_size // self.patch_size) ** 2
snake_case: Union[str, Any] = self.patch_size**2 * self.num_channels
self.parent.assertEqual(result.logits.shape , (self.batch_size, num_patches, expected_num_channels) )
# test greyscale images
snake_case: Optional[int] = 1
snake_case: Union[str, Any] = TFViTMAEForPreTraining(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
snake_case: int = model(SCREAMING_SNAKE_CASE__ , training=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.patch_size**2
self.parent.assertEqual(result.logits.shape , (self.batch_size, num_patches, expected_num_channels) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.prepare_config_and_inputs()
((snake_case) , (snake_case) , (snake_case)): int = config_and_inputs
snake_case: str = {'pixel_values': pixel_values}
return config, inputs_dict
@require_tf
class SCREAMING_SNAKE_CASE ( snake_case , snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = (TFViTMAEModel, TFViTMAEForPreTraining) if is_tf_available() else ()
__UpperCamelCase = {"feature-extraction": TFViTMAEModel} if is_tf_available() else {}
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = TFViTMAEModelTester(self )
snake_case: Dict = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE__ , has_text_modality=SCREAMING_SNAKE_CASE__ , hidden_size=37 )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.config_tester.run_common_tests()
@unittest.skip(reason='ViTMAE does not use inputs_embeds' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Any = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case: Optional[int] = model_class(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(model.get_input_embeddings() , (tf.keras.layers.Layer) )
snake_case: Any = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(SCREAMING_SNAKE_CASE__ , tf.keras.layers.Layer ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case , snake_case: Optional[int] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case: Union[str, Any] = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = inspect.signature(model.call )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
snake_case: Union[str, Any] = [*signature.parameters.keys()]
snake_case: List[str] = ['pixel_values']
self.assertListEqual(arg_names[:1] , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_pretraining(*SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case , snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[str] = int((config.image_size // config.patch_size) ** 2 )
snake_case: List[str] = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
for model_class in self.all_model_classes:
snake_case: Optional[Any] = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = copy.deepcopy(self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
snake_case: Optional[Any] = model(**SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = outputs_dict[0].numpy()
snake_case: Optional[int] = outputs_keywords[0].numpy()
self.assertLess(np.sum(np.abs(output_dict - output_keywords ) ) , 1E-6 )
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case , snake_case: Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[Any] = int((config.image_size // config.patch_size) ** 2 )
snake_case: int = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
def prepare_numpy_arrays(SCREAMING_SNAKE_CASE__ ):
snake_case: List[str] = {}
for k, v in inputs_dict.items():
if tf.is_tensor(SCREAMING_SNAKE_CASE__ ):
snake_case: Optional[Any] = v.numpy()
else:
snake_case: Tuple = np.array(SCREAMING_SNAKE_CASE__ )
return inputs_np_dict
for model_class in self.all_model_classes:
snake_case: Optional[int] = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: str = self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = prepare_numpy_arrays(SCREAMING_SNAKE_CASE__ )
snake_case: int = model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = model(**SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
self.assert_outputs_same(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
np.random.seed(2 )
snake_case: Optional[Any] = int((tf_model.config.image_size // tf_model.config.patch_size) ** 2 )
snake_case: Any = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
# Add `noise` argument.
# PT inputs will be prepared in `super().check_pt_tf_models()` with this added `noise` argument
snake_case: Tuple = tf_noise
super().check_pt_tf_models(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case , snake_case: int = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: List[str] = {
module_member
for model_class in self.all_model_classes
for module in (import_module(model_class.__module__ ),)
for module_member_name in dir(SCREAMING_SNAKE_CASE__ )
if module_member_name.endswith('MainLayer' )
# This condition is required, since `modeling_tf_clip.py` has 3 classes whose names end with `MainLayer`.
and module_member_name[: -len('MainLayer' )] == model_class.__name__[: -len('Model' )]
for module_member in (getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ),)
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
and tf.keras.layers.Layer in module_member.__bases__
and getattr(SCREAMING_SNAKE_CASE__ , '_keras_serializable' , SCREAMING_SNAKE_CASE__ )
}
snake_case: List[str] = int((config.image_size // config.patch_size) ** 2 )
snake_case: str = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
snake_case: Tuple = tf.convert_to_tensor(SCREAMING_SNAKE_CASE__ )
inputs_dict.update({'noise': noise} )
for main_layer_class in tf_main_layer_classes:
snake_case: Optional[int] = main_layer_class(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = {
name: tf.keras.Input(tensor.shape[1:] , dtype=tensor.dtype ) for name, tensor in inputs_dict.items()
}
snake_case: List[str] = tf.keras.Model(SCREAMING_SNAKE_CASE__ , outputs=main_layer(SCREAMING_SNAKE_CASE__ ) )
snake_case: Optional[Any] = model(SCREAMING_SNAKE_CASE__ )
with tempfile.TemporaryDirectory() as tmpdirname:
snake_case: Optional[int] = os.path.join(SCREAMING_SNAKE_CASE__ , 'keras_model.h5' )
model.save(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tf.keras.models.load_model(
SCREAMING_SNAKE_CASE__ , custom_objects={main_layer_class.__name__: main_layer_class} )
assert isinstance(SCREAMING_SNAKE_CASE__ , tf.keras.Model )
snake_case: Any = model(SCREAMING_SNAKE_CASE__ )
self.assert_outputs_same(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case , snake_case: List[str] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: Optional[int] = int((config.image_size // config.patch_size) ** 2 )
snake_case: Union[str, Any] = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
for model_class in self.all_model_classes:
snake_case: Dict = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
if model_class.__name__ == "TFViTMAEModel":
snake_case: int = outputs.last_hidden_state.numpy()
snake_case: Any = 0
else:
snake_case: Any = outputs.logits.numpy()
snake_case: Union[str, Any] = 0
with tempfile.TemporaryDirectory() as tmpdirname:
model.save_pretrained(SCREAMING_SNAKE_CASE__ , saved_model=SCREAMING_SNAKE_CASE__ )
snake_case: str = model_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
if model_class.__name__ == "TFViTMAEModel":
snake_case: List[str] = after_outputs['last_hidden_state'].numpy()
snake_case: Optional[Any] = 0
else:
snake_case: Union[str, Any] = after_outputs['logits'].numpy()
snake_case: List[Any] = 0
snake_case: Optional[Any] = np.amax(np.abs(out_a - out_a ) )
self.assertLessEqual(SCREAMING_SNAKE_CASE__ , 1E-5 )
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case , snake_case: List[str] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case: str = int((config.image_size // config.patch_size) ** 2 )
snake_case: Union[str, Any] = np.random.uniform(size=(self.model_tester.batch_size, num_patches) )
for model_class in self.all_model_classes:
snake_case: Optional[Any] = model_class(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self._prepare_for_class(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = model.get_config()
# make sure that returned config is jsonifiable, which is required by keras
json.dumps(SCREAMING_SNAKE_CASE__ )
snake_case: str = model_class.from_config(model.get_config() )
# make sure it also accepts a normal config
snake_case: str = model_class.from_config(model.config )
snake_case: Union[str, Any] = new_model(SCREAMING_SNAKE_CASE__ ) # Build model
new_model.set_weights(model.get_weights() )
snake_case: Optional[Any] = new_model(SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
self.assert_outputs_same(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@unittest.skip(
reason='ViTMAE returns a random mask + ids_restore in each forward pass. See test_save_load\n to get deterministic results.' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@unittest.skip(reason='ViTMAE returns a random mask + ids_restore in each forward pass. See test_save_load' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = TFViTMAEModel.from_pretrained('google/vit-base-patch16-224' )
self.assertIsNotNone(SCREAMING_SNAKE_CASE__ )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Optional[int] = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' )
return image
@require_tf
@require_vision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ViTImageProcessor.from_pretrained('facebook/vit-mae-base' ) if is_vision_available() else None
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
np.random.seed(2 )
snake_case: int = TFViTMAEForPreTraining.from_pretrained('facebook/vit-mae-base' )
snake_case: Dict = self.default_image_processor
snake_case: Optional[Any] = prepare_img()
snake_case: int = image_processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='tf' )
# prepare a noise vector that will be also used for testing the TF model
# (this way we can ensure that the PT and TF models operate on the same inputs)
snake_case: Optional[int] = ViTMAEConfig()
snake_case: Any = int((vit_mae_config.image_size // vit_mae_config.patch_size) ** 2 )
snake_case: int = np.random.uniform(size=(1, num_patches) )
# forward pass
snake_case: Union[str, Any] = model(**SCREAMING_SNAKE_CASE__ , noise=SCREAMING_SNAKE_CASE__ )
# verify the logits
snake_case: List[Any] = tf.convert_to_tensor([1, 1_96, 7_68] )
self.assertEqual(outputs.logits.shape , SCREAMING_SNAKE_CASE__ )
snake_case: int = tf.convert_to_tensor(
[[-0.05_48, -1.70_23, -0.93_25], [0.37_21, -0.56_70, -0.22_33], [0.82_35, -1.38_78, -0.35_24]] )
tf.debugging.assert_near(outputs.logits[0, :3, :3] , SCREAMING_SNAKE_CASE__ , atol=1E-4 ) | 692 |
'''simple docstring'''
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCAmelCase = "pt"
elif is_tf_available():
__UpperCAmelCase = "tf"
else:
__UpperCAmelCase = "jax"
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ByTaTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: int = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ByTaTokenizer.from_pretrained('google/byt5-small' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=20 , SCREAMING_SNAKE_CASE__=5 ):
'''simple docstring'''
snake_case: Optional[Any] = []
for i in range(len(SCREAMING_SNAKE_CASE__ ) ):
try:
snake_case: Optional[Any] = tokenizer.decode([i] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
except UnicodeDecodeError:
pass
toks.append((i, tok) )
snake_case: List[str] = list(filter(lambda SCREAMING_SNAKE_CASE__ : re.match(r'^[ a-zA-Z]+$' , t[1] ) , SCREAMING_SNAKE_CASE__ ) )
snake_case: str = list(filter(lambda SCREAMING_SNAKE_CASE__ : [t[0]] == tokenizer.encode(t[1] , add_special_tokens=SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ ) )
if max_length is not None and len(SCREAMING_SNAKE_CASE__ ) > max_length:
snake_case: Union[str, Any] = toks[:max_length]
if min_length is not None and len(SCREAMING_SNAKE_CASE__ ) < min_length and len(SCREAMING_SNAKE_CASE__ ) > 0:
while len(SCREAMING_SNAKE_CASE__ ) < min_length:
snake_case: Tuple = toks + toks
# toks_str = [t[1] for t in toks]
snake_case: Dict = [t[0] for t in toks]
# Ensure consistency
snake_case: int = tokenizer.decode(SCREAMING_SNAKE_CASE__ , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
if " " not in output_txt and len(SCREAMING_SNAKE_CASE__ ) > 1:
snake_case: str = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
)
if with_prefix_space:
snake_case: Tuple = ' ' + output_txt
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
return output_txt, output_ids
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: str = tokenizer(['hi</s>', 'I went to the gym</s>', '</s>'] )
snake_case: List[Any] = tokenizer(['hi', 'I went to the gym', ''] )
self.assertListEqual(batch_with_eos_added['input_ids'] , batch_without_eos_added['input_ids'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: Union[str, Any] = 'Unicode €.'
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [88, 1_13, 1_08, 1_02, 1_14, 1_03, 1_04, 35, 2_29, 1_33, 1_75, 49, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[str] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'Unicode €.</s>' )
snake_case: List[Any] = tokenizer('e è é ê ë' )
snake_case: Optional[Any] = [1_04, 35, 1_98, 1_71, 35, 1_98, 1_72, 35, 1_98, 1_73, 35, 1_98, 1_74, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[Any] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'e è é ê ë</s>' )
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë' ) ) , 'e è é ê ë</s>' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.ta_base_tokenizer
snake_case: Optional[Any] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case: Optional[int] = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 1, 0]
# fmt: on
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if FRAMEWORK != "jax":
snake_case: Optional[Any] = list(batch.input_ids.numpy()[0] )
else:
snake_case: Dict = list(batch.input_ids.tolist()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 37) , batch.input_ids.shape )
self.assertEqual((2, 37) , batch.attention_mask.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.ta_base_tokenizer
snake_case: List[str] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case: Union[str, Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_attention_mask' , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.ta_base_tokenizer
snake_case: str = [
'Summary of the text.',
'Another summary.',
]
snake_case: Dict = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.ta_base_tokenizer
snake_case: Optional[int] = ['A long paragraph for summarization. </s>']
snake_case: str = ['Summary of the text. </s>']
# fmt: off
snake_case: str = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 35, 1]
snake_case: Optional[int] = [86, 1_20, 1_12, 1_12, 1_00, 1_17, 1_24, 35, 1_14, 1_05, 35, 1_19, 1_07, 1_04, 35, 1_19, 1_04, 1_23, 1_19, 49, 35, 1]
# fmt: on
snake_case: List[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , text_target=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['input_ids'][0] )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['labels'][0] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: Optional[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: Union[str, Any] = tempfile.mkdtemp()
snake_case: Dict = ' He is very happy, UNwant\u00E9d,running'
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'] )
snake_case: List[str] = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token' )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: int = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn('new_additional_special_token' , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Union[str, Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: str = json.load(SCREAMING_SNAKE_CASE__ )
snake_case: int = [F"""<extra_id_{i}>""" for i in range(1_25 )]
snake_case: Optional[int] = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case: str = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Dict = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , )
self.assertIn(
'an_additional_special_token' , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['an_additional_special_token'] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'] ) ) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: Union[str, Any] = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token' , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: Union[str, Any] = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , )
self.assertIn('a_new_additional_special_token' , tokenizer.additional_special_tokens )
self.assertEqual(
['a_new_additional_special_token'] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'] ) ) , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertTrue(tokenizer.decode([2_55] ) == '' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_tokenizers(fast=SCREAMING_SNAKE_CASE__ , do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = ['t', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 'x', 't', '</s>']
snake_case: List[str] = tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Optional[Any] = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Dict = 0
snake_case: List[Any] = tokenizer.convert_ids_to_tokens(
SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [token_id_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [token_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [token_id_to_test_setters] ) | 692 | 1 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
from typing import Any, Dict, Optional
import torch
import torch.nn.functional as F
from torch import nn
from ..utils import maybe_allow_in_graph
from .activations import get_activation
from .attention_processor import Attention
from .embeddings import CombinedTimestepLabelEmbeddings
@maybe_allow_in_graph
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "layer_norm" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = only_cross_attention
snake_case: Optional[Any] = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm_zero'
snake_case: Tuple = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm'
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
raise ValueError(
F"""`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"""
F""" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}.""" )
# Define 3 blocks. Each block has its own normalization layer.
# 1. Self-Attn
if self.use_ada_layer_norm:
snake_case: List[str] = AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case: str = AdaLayerNormZero(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
else:
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if only_cross_attention else None , upcast_attention=SCREAMING_SNAKE_CASE__ , )
# 2. Cross-Attn
if cross_attention_dim is not None or double_self_attention:
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
# the second cross attention block.
snake_case: Tuple = (
AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm
else nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if not double_self_attention else None , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , upcast_attention=SCREAMING_SNAKE_CASE__ , ) # is self-attn if encoder_hidden_states is none
else:
snake_case: int = None
snake_case: Tuple = None
# 3. Feed-forward
snake_case: Union[str, Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = FeedForward(SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , activation_fn=SCREAMING_SNAKE_CASE__ , final_dropout=SCREAMING_SNAKE_CASE__ )
# let chunk size default to None
snake_case: Any = None
snake_case: Any = 0
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = chunk_size
snake_case: str = dim
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
if self.use_ada_layer_norm:
snake_case: Optional[int] = self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case , snake_case , snake_case , snake_case , snake_case: int = self.norma(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=hidden_states.dtype )
else:
snake_case: List[str] = self.norma(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = cross_attention_kwargs if cross_attention_kwargs is not None else {}
snake_case: List[str] = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=encoder_hidden_states if self.only_cross_attention else None , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if self.use_ada_layer_norm_zero:
snake_case: Tuple = gate_msa.unsqueeze(1 ) * attn_output
snake_case: List[str] = attn_output + hidden_states
# 2. Cross-Attention
if self.attna is not None:
snake_case: Dict = (
self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) if self.use_ada_layer_norm else self.norma(SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: List[str] = attn_output + hidden_states
# 3. Feed-forward
snake_case: str = self.norma(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: str = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
if self._chunk_size is not None:
# "feed_forward_chunk_size" can be used to save memory
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
raise ValueError(
F"""`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`.""" )
snake_case: List[str] = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
snake_case: Optional[Any] = torch.cat(
[self.ff(SCREAMING_SNAKE_CASE__ ) for hid_slice in norm_hidden_states.chunk(SCREAMING_SNAKE_CASE__ , dim=self._chunk_dim )] , dim=self._chunk_dim , )
else:
snake_case: int = self.ff(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: Union[str, Any] = gate_mlp.unsqueeze(1 ) * ff_output
snake_case: Tuple = ff_output + hidden_states
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 4 , SCREAMING_SNAKE_CASE__ = 0.0 , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: int = int(dim * mult )
snake_case: Optional[Any] = dim_out if dim_out is not None else dim
if activation_fn == "gelu":
snake_case: int = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if activation_fn == "gelu-approximate":
snake_case: Optional[Any] = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , approximate='tanh' )
elif activation_fn == "geglu":
snake_case: List[Any] = GEGLU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif activation_fn == "geglu-approximate":
snake_case: Optional[int] = ApproximateGELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.ModuleList([] )
# project in
self.net.append(SCREAMING_SNAKE_CASE__ )
# project dropout
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
# project out
self.net.append(nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
if final_dropout:
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for module in self.net:
snake_case: Optional[int] = module(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "none" ):
'''simple docstring'''
super().__init__()
snake_case: Optional[int] = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = approximate
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ , approximate=self.approximate )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) , approximate=self.approximate ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.proj(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.gelu(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = nn.Linear(SCREAMING_SNAKE_CASE__ , dim_out * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: int = self.proj(SCREAMING_SNAKE_CASE__ ).chunk(2 , dim=-1 )
return hidden_states * self.gelu(SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = self.proj(SCREAMING_SNAKE_CASE__ )
return x * torch.sigmoid(1.7_02 * x )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Optional[Any] = nn.Embedding(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = nn.SiLU()
snake_case: Union[str, Any] = nn.Linear(SCREAMING_SNAKE_CASE__ , embedding_dim * 2 )
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case: Dict = torch.chunk(SCREAMING_SNAKE_CASE__ , 2 )
snake_case: str = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale) + shift
return x
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = CombinedTimestepLabelEmbeddings(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.SiLU()
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , 6 * embedding_dim , bias=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ , eps=1E-6 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
snake_case: int = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case , snake_case , snake_case , snake_case , snake_case: str = emb.chunk(6 , dim=1 )
snake_case: Dict = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale_msa[:, None]) + shift_msa[:, None]
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 1E-5 ):
'''simple docstring'''
super().__init__()
snake_case: str = num_groups
snake_case: str = eps
if act_fn is None:
snake_case: Dict = None
else:
snake_case: List[str] = get_activation(SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , out_dim * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.act:
snake_case: Optional[Any] = self.act(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.linear(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = emb[:, :, None, None]
snake_case , snake_case: List[Any] = emb.chunk(2 , dim=1 )
snake_case: Any = F.group_norm(SCREAMING_SNAKE_CASE__ , self.num_groups , eps=self.eps )
snake_case: Optional[int] = x * (1 + scale) + shift
return x | 692 | 1 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
from sklearn.metrics import mean_squared_error
import datasets
__UpperCAmelCase = "\\n@article{scikit-learn,\n title={Scikit-learn: Machine Learning in {P}ython},\n author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V.\n and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P.\n and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and\n Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.},\n journal={Journal of Machine Learning Research},\n volume={12},\n pages={2825--2830},\n year={2011}\n}\n"
__UpperCAmelCase = "\\nMean Squared Error(MSE) is the average of the square of difference between the predicted\nand actual values.\n"
__UpperCAmelCase = "\nArgs:\n predictions: array-like of shape (n_samples,) or (n_samples, n_outputs)\n Estimated target values.\n references: array-like of shape (n_samples,) or (n_samples, n_outputs)\n Ground truth (correct) target values.\n sample_weight: array-like of shape (n_samples,), default=None\n Sample weights.\n multioutput: {\"raw_values\", \"uniform_average\"} or array-like of shape (n_outputs,), default=\"uniform_average\"\n Defines aggregating of multiple output values. Array-like value defines weights used to average errors.\n\n \"raw_values\" : Returns a full set of errors in case of multioutput input.\n\n \"uniform_average\" : Errors of all outputs are averaged with uniform weight.\n\n squared : bool, default=True\n If True returns MSE value, if False returns RMSE (Root Mean Squared Error) value.\n\nReturns:\n mse : mean squared error.\nExamples:\n\n >>> mse_metric = datasets.load_metric(\"mse\")\n >>> predictions = [2.5, 0.0, 2, 8]\n >>> references = [3, -0.5, 2, 7]\n >>> results = mse_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {'mse': 0.375}\n >>> rmse_result = mse_metric.compute(predictions=predictions, references=references, squared=False)\n >>> print(rmse_result)\n {'mse': 0.6123724356957945}\n\n If you're using multi-dimensional lists, then set the config as follows :\n\n >>> mse_metric = datasets.load_metric(\"mse\", \"multilist\")\n >>> predictions = [[0.5, 1], [-1, 1], [7, -6]]\n >>> references = [[0, 2], [-1, 2], [8, -5]]\n >>> results = mse_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {'mse': 0.7083333333333334}\n >>> results = mse_metric.compute(predictions=predictions, references=references, multioutput='raw_values')\n >>> print(results) # doctest: +NORMALIZE_WHITESPACE\n {'mse': array([0.41666667, 1. ])}\n"
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class SCREAMING_SNAKE_CASE ( datasets.Metric ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(self._get_feature_types() ) , reference_urls=[
'https://scikit-learn.org/stable/modules/generated/sklearn.metrics.mean_squared_error.html'
] , )
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.config_name == "multilist":
return {
"predictions": datasets.Sequence(datasets.Value('float' ) ),
"references": datasets.Sequence(datasets.Value('float' ) ),
}
else:
return {
"predictions": datasets.Value('float' ),
"references": datasets.Value('float' ),
}
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__="uniform_average" , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: Optional[Any] = mean_squared_error(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , sample_weight=SCREAMING_SNAKE_CASE__ , multioutput=SCREAMING_SNAKE_CASE__ , squared=SCREAMING_SNAKE_CASE__ )
return {"mse": mse} | 692 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 3 , __A : int = 7 , __A : int = 1_00_00_00 ):
'''simple docstring'''
snake_case: Optional[int] = 0
snake_case: Optional[int] = 1
for current_denominator in range(1 , limit + 1 ):
snake_case: List[Any] = current_denominator * numerator // denominator
if current_denominator % denominator == 0:
current_numerator -= 1
if current_numerator * max_denominator > current_denominator * max_numerator:
snake_case: Any = current_numerator
snake_case: str = current_denominator
return max_numerator
if __name__ == "__main__":
print(solution(numerator=3, denominator=7, limit=1_000_000)) | 692 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_roformer": ["ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoFormerConfig", "RoFormerOnnxConfig"],
"tokenization_roformer": ["RoFormerTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["RoFormerTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoFormerForCausalLM",
"RoFormerForMaskedLM",
"RoFormerForMultipleChoice",
"RoFormerForQuestionAnswering",
"RoFormerForSequenceClassification",
"RoFormerForTokenClassification",
"RoFormerLayer",
"RoFormerModel",
"RoFormerPreTrainedModel",
"load_tf_weights_in_roformer",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFRoFormerForCausalLM",
"TFRoFormerForMaskedLM",
"TFRoFormerForMultipleChoice",
"TFRoFormerForQuestionAnswering",
"TFRoFormerForSequenceClassification",
"TFRoFormerForTokenClassification",
"TFRoFormerLayer",
"TFRoFormerModel",
"TFRoFormerPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"FlaxRoFormerForMaskedLM",
"FlaxRoFormerForMultipleChoice",
"FlaxRoFormerForQuestionAnswering",
"FlaxRoFormerForSequenceClassification",
"FlaxRoFormerForTokenClassification",
"FlaxRoFormerModel",
"FlaxRoFormerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig
from .tokenization_roformer import RoFormerTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_roformer_fast import RoFormerTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roformer import (
ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
RoFormerForCausalLM,
RoFormerForMaskedLM,
RoFormerForMultipleChoice,
RoFormerForQuestionAnswering,
RoFormerForSequenceClassification,
RoFormerForTokenClassification,
RoFormerLayer,
RoFormerModel,
RoFormerPreTrainedModel,
load_tf_weights_in_roformer,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_roformer import (
TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TFRoFormerForCausalLM,
TFRoFormerForMaskedLM,
TFRoFormerForMultipleChoice,
TFRoFormerForQuestionAnswering,
TFRoFormerForSequenceClassification,
TFRoFormerForTokenClassification,
TFRoFormerLayer,
TFRoFormerModel,
TFRoFormerPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_roformer import (
FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
FlaxRoFormerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 | 1 |
'''simple docstring'''
import warnings
from ...utils import logging
from .image_processing_deformable_detr import DeformableDetrImageProcessor
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
warnings.warn(
'The class DeformableDetrFeatureExtractor is deprecated and will be removed in version 5 of Transformers.'
' Please use DeformableDetrImageProcessor instead.' , SCREAMING_SNAKE_CASE__ , )
super().__init__(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 | 1 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ...utils.backbone_utils import BackboneConfigMixin, get_aligned_output_features_output_indices
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/focalnet-tiny": "https://huggingface.co/microsoft/focalnet-tiny/resolve/main/config.json",
}
class SCREAMING_SNAKE_CASE ( snake_case , snake_case ):
'''simple docstring'''
__UpperCamelCase = "focalnet"
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=96 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=[1_92, 3_84, 7_68, 7_68] , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[2, 2, 2, 2] , SCREAMING_SNAKE_CASE__=[3, 3, 3, 3] , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=1E-4 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = image_size
snake_case: Tuple = patch_size
snake_case: str = num_channels
snake_case: Dict = embed_dim
snake_case: str = use_conv_embed
snake_case: Tuple = hidden_sizes
snake_case: int = depths
snake_case: List[str] = focal_levels
snake_case: Any = focal_windows
snake_case: Any = hidden_act
snake_case: Optional[int] = mlp_ratio
snake_case: List[str] = hidden_dropout_prob
snake_case: Optional[Any] = drop_path_rate
snake_case: Dict = use_layerscale
snake_case: Union[str, Any] = layerscale_value
snake_case: List[Any] = use_post_layernorm
snake_case: int = use_post_layernorm_in_modulation
snake_case: Optional[Any] = normalize_modulator
snake_case: int = initializer_range
snake_case: Union[str, Any] = layer_norm_eps
snake_case: Optional[int] = encoder_stride
snake_case: Any = ['stem'] + [F"""stage{idx}""" for idx in range(1 , len(self.depths ) + 1 )]
snake_case , snake_case: List[str] = get_aligned_output_features_output_indices(
out_features=SCREAMING_SNAKE_CASE__ , out_indices=SCREAMING_SNAKE_CASE__ , stage_names=self.stage_names ) | 692 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : Any=2_81_23 ):
'''simple docstring'''
snake_case: Optional[int] = [1] * (limit + 1)
for i in range(2 , int(limit**0.5 ) + 1 ):
sum_divs[i * i] += i
for k in range(i + 1 , limit // i + 1 ):
sum_divs[k * i] += k + i
snake_case: Dict = set()
snake_case: Tuple = 0
for n in range(1 , limit + 1 ):
if sum_divs[n] > n:
abundants.add(__A )
if not any((n - a in abundants) for a in abundants ):
res += n
return res
if __name__ == "__main__":
print(solution()) | 692 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available
__UpperCAmelCase = {
"configuration_rag": ["RagConfig"],
"retrieval_rag": ["RagRetriever"],
"tokenization_rag": ["RagTokenizer"],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"RagModel",
"RagPreTrainedModel",
"RagSequenceForGeneration",
"RagTokenForGeneration",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TFRagModel",
"TFRagPreTrainedModel",
"TFRagSequenceForGeneration",
"TFRagTokenForGeneration",
]
if TYPE_CHECKING:
from .configuration_rag import RagConfig
from .retrieval_rag import RagRetriever
from .tokenization_rag import RagTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_rag import RagModel, RagPreTrainedModel, RagSequenceForGeneration, RagTokenForGeneration
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_rag import (
TFRagModel,
TFRagPreTrainedModel,
TFRagSequenceForGeneration,
TFRagTokenForGeneration,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 1 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
from transformers import BatchEncoding, CanineTokenizer
from transformers.testing_utils import require_tokenizers, require_torch
from transformers.tokenization_utils import AddedToken
from transformers.utils import cached_property
from ...test_tokenization_common import TokenizerTesterMixin
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = CanineTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: List[Any] = CanineTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return CanineTokenizer.from_pretrained('google/canine-s' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
snake_case: Any = 10_24
return tokenizer
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.canine_tokenizer
snake_case: Union[str, Any] = ['Life is like a box of chocolates.', 'You never know what you\'re gonna get.']
# fmt: off
snake_case: Any = [5_73_44, 76, 1_05, 1_02, 1_01, 32, 1_05, 1_15, 32, 1_08, 1_05, 1_07, 1_01, 32, 97, 32, 98, 1_11, 1_20, 32, 1_11, 1_02, 32, 99, 1_04, 1_11, 99, 1_11, 1_08, 97, 1_16, 1_01, 1_15, 46, 5_73_45, 0, 0, 0, 0]
# fmt: on
snake_case: List[str] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors='pt' )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = list(batch.input_ids.numpy()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 39) , batch.input_ids.shape )
self.assertEqual((2, 39) , batch.attention_mask.shape )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.canine_tokenizer
snake_case: Optional[Any] = ['Once there was a man.', 'He wrote a test in HuggingFace Tranformers.']
snake_case: Optional[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors='pt' )
# check if input_ids, attention_mask and token_type_ids are returned
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertIn('token_type_ids' , SCREAMING_SNAKE_CASE__ )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.canine_tokenizer
snake_case: List[str] = [
'What\'s the weater?',
'It\'s about 25 degrees.',
]
snake_case: int = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors='pt' )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: str = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: Union[str, Any] = ' He is very happy, UNwant\u00E9d,running'
snake_case: Dict = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: int = tempfile.mkdtemp()
snake_case: Tuple = ' He is very happy, UNwant\u00E9d,running'
snake_case: str = tokenizer.additional_special_tokens
# We can add a new special token for Canine as follows:
snake_case: Optional[int] = chr(0xe_007 )
additional_special_tokens.append(SCREAMING_SNAKE_CASE__ )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: Union[str, Any] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn(SCREAMING_SNAKE_CASE__ , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case , snake_case: Any = self.get_clean_sequence(SCREAMING_SNAKE_CASE__ )
# a special token for Canine can be defined as follows:
snake_case: Optional[int] = 0xe_005
snake_case: List[str] = chr(SCREAMING_SNAKE_CASE__ )
tokenizer.add_special_tokens({'cls_token': special_token} )
snake_case: Dict = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , 1 )
snake_case: List[Any] = tokenizer.decode(ids + encoded_special_token , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , input_encoded + special_token_id )
snake_case: Tuple = tokenizer.decode(SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertTrue(special_token not in decoded )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = chr(0xe_005 )
snake_case: Optional[int] = chr(0xe_006 )
# `add_tokens` method stores special tokens only in `tokenizer.unique_no_split_tokens`. (in tokenization_utils.py)
tokenizer.add_tokens([SPECIAL_TOKEN_1] , special_tokens=SCREAMING_SNAKE_CASE__ )
# `add_special_tokens` method stores special tokens in `tokenizer.additional_special_tokens`,
# which also occur in `tokenizer.all_special_tokens`. (in tokenization_utils_base.py)
tokenizer.add_special_tokens({'additional_special_tokens': [SPECIAL_TOKEN_2]} )
snake_case: Optional[Any] = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , 1 )
self.assertEqual(len(SCREAMING_SNAKE_CASE__ ) , 1 )
self.assertEqual(token_a[0] , SCREAMING_SNAKE_CASE__ )
self.assertEqual(token_a[0] , SCREAMING_SNAKE_CASE__ )
@require_tokenizers
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# a special token for Canine can be defined as follows:
snake_case: Union[str, Any] = 0xe_006
snake_case: Tuple = chr(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = AddedToken(SCREAMING_SNAKE_CASE__ , lstrip=SCREAMING_SNAKE_CASE__ )
tokenizer.add_special_tokens({'additional_special_tokens': [new_token]} )
with tempfile.TemporaryDirectory() as tmp_dir_name:
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
tokenizer.from_pretrained(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: int = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
# a special token for Canine can be defined as follows:
snake_case: Union[str, Any] = 0xe_006
snake_case: int = chr(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = [new_token_a]
snake_case: Dict = [new_token_a]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Tuple = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , extra_ids=0 )
self.assertIn(SCREAMING_SNAKE_CASE__ , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
[new_token_a] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids([new_token_a] ) ) , )
snake_case: Optional[Any] = 0xe_007
snake_case: Dict = chr(SCREAMING_SNAKE_CASE__ )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: int = [AddedToken(SCREAMING_SNAKE_CASE__ , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: str = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , extra_ids=0 )
self.assertIn(SCREAMING_SNAKE_CASE__ , tokenizer.additional_special_tokens )
# self.assertIn(new_token_2,tokenizer.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
[new_token_a] , tokenizer.convert_ids_to_tokens(tokenizer.convert_tokens_to_ids([new_token_a] ) ) )
@require_tokenizers
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: List[Any] = 'hello world'
if self.space_between_special_tokens:
snake_case: List[str] = '[CLS] hello world [SEP]'
else:
snake_case: Dict = input
snake_case: Union[str, Any] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.decode(SCREAMING_SNAKE_CASE__ , spaces_between_special_tokens=self.space_between_special_tokens )
self.assertIn(SCREAMING_SNAKE_CASE__ , [output, output.lower()] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: str = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Optional[Any] = 'a'
snake_case: List[str] = ord(SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
snake_case: List[Any] = 0xe_006
snake_case: List[Any] = chr(SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [additional_special_token_id] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [additional_special_token] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [additional_special_token_id] )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass | 692 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = "▁"
__UpperCAmelCase = {"vocab_file": "sentencepiece.bpe.model"}
__UpperCAmelCase = {
"vocab_file": {
"facebook/xglm-564M": "https://huggingface.co/facebook/xglm-564M/resolve/main/sentencepiece.bpe.model",
}
}
__UpperCAmelCase = {
"facebook/xglm-564M": 2_048,
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
# Compatibility with the original tokenizer
snake_case: Optional[Any] = 7
snake_case: List[str] = [F"""<madeupword{i}>""" for i in range(self.num_madeup_words )]
snake_case: str = kwargs.get('additional_special_tokens' , [] )
kwargs["additional_special_tokens"] += [
word for word in madeup_words if word not in kwargs["additional_special_tokens"]
]
super().__init__(
bos_token=SCREAMING_SNAKE_CASE__ , eos_token=SCREAMING_SNAKE_CASE__ , unk_token=SCREAMING_SNAKE_CASE__ , sep_token=SCREAMING_SNAKE_CASE__ , cls_token=SCREAMING_SNAKE_CASE__ , pad_token=SCREAMING_SNAKE_CASE__ , sp_model_kwargs=self.sp_model_kwargs , **SCREAMING_SNAKE_CASE__ , )
snake_case: int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(SCREAMING_SNAKE_CASE__ ) )
snake_case: int = vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
snake_case: Tuple = 1
# Mimic fairseq token-to-id alignment for the first 4 token
snake_case: Optional[Any] = {'<s>': 0, '<pad>': 1, '</s>': 2, '<unk>': 3}
snake_case: Union[str, Any] = len(self.sp_model )
snake_case: str = {F"""<madeupword{i}>""": sp_size + i + self.fairseq_offset for i in range(self.num_madeup_words )}
self.fairseq_tokens_to_ids.update(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def __getstate__( self ):
'''simple docstring'''
snake_case: List[Any] = self.__dict__.copy()
snake_case: Union[str, Any] = None
snake_case: Union[str, Any] = self.sp_model.serialized_model_proto()
return state
def __setstate__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
snake_case: Union[str, Any] = {}
snake_case: Tuple = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.LoadFromSerializedProto(self.sp_model_proto )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.sep_token_id] + token_ids_a
snake_case: Optional[Any] = [self.sep_token_id]
return sep + token_ids_a + sep + sep + token_ids_a
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=SCREAMING_SNAKE_CASE__ , token_ids_a=SCREAMING_SNAKE_CASE__ , already_has_special_tokens=SCREAMING_SNAKE_CASE__ )
if token_ids_a is None:
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ )) + [1, 1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
if token_ids_a is None:
return len(sep + token_ids_a ) * [0]
return len(sep + token_ids_a + sep + sep + token_ids_a ) * [0]
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.sp_model ) + self.fairseq_offset + self.num_madeup_words
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = {self.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.sp_model.encode(SCREAMING_SNAKE_CASE__ , out_type=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
snake_case: Dict = self.sp_model.PieceToId(SCREAMING_SNAKE_CASE__ )
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = ''.join(SCREAMING_SNAKE_CASE__ ).replace(SCREAMING_SNAKE_CASE__ , ' ' ).strip()
return out_string
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(SCREAMING_SNAKE_CASE__ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: List[str] = os.path.join(
SCREAMING_SNAKE_CASE__ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(SCREAMING_SNAKE_CASE__ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , SCREAMING_SNAKE_CASE__ )
elif not os.path.isfile(self.vocab_file ):
with open(SCREAMING_SNAKE_CASE__ , 'wb' ) as fi:
snake_case: int = self.sp_model.serialized_model_proto()
fi.write(SCREAMING_SNAKE_CASE__ )
return (out_vocab_file,) | 692 | 1 |
'''simple docstring'''
import datasets
import faiss
import numpy as np
import streamlit as st
import torch
from elasticsearch import Elasticsearch
from elia_utils import (
embed_questions_for_retrieval,
make_qa_sas_model,
qa_sas_generate,
query_es_index,
query_qa_dense_index,
)
import transformers
from transformers import AutoModel, AutoModelForSeqaSeqLM, AutoTokenizer
__UpperCAmelCase = "bart"
__UpperCAmelCase = True
@st.cache(allow_output_mutation=__A )
def lowerCAmelCase_ ( ):
'''simple docstring'''
if LOAD_DENSE_INDEX:
snake_case: Any = AutoTokenizer.from_pretrained('yjernite/retribert-base-uncased' )
snake_case: Any = AutoModel.from_pretrained('yjernite/retribert-base-uncased' ).to('cuda:0' )
snake_case: List[str] = qar_model.eval()
else:
snake_case , snake_case: Dict = (None, None)
if MODEL_TYPE == "bart":
snake_case: List[Any] = AutoTokenizer.from_pretrained('yjernite/bart_eli5' )
snake_case: Optional[int] = AutoModelForSeqaSeqLM.from_pretrained('yjernite/bart_eli5' ).to('cuda:0' )
snake_case: str = torch.load('seq2seq_models/eli5_bart_model_blm_2.pth' )
sas_model.load_state_dict(save_dict['model'] )
snake_case: Dict = sas_model.eval()
else:
snake_case , snake_case: str = make_qa_sas_model(
model_name='t5-small' , from_file='seq2seq_models/eli5_t5_model_1024_4.pth' , device='cuda:0' )
return (qar_tokenizer, qar_model, sas_tokenizer, sas_model)
@st.cache(allow_output_mutation=__A )
def lowerCAmelCase_ ( ):
'''simple docstring'''
if LOAD_DENSE_INDEX:
snake_case: Tuple = faiss.StandardGpuResources()
snake_case: List[Any] = datasets.load_dataset(path='wiki_snippets' , name='wiki40b_en_100_0' )['train']
snake_case: int = np.memmap(
'wiki40b_passages_reps_32_l-8_h-768_b-512-512.dat' , dtype='float32' , mode='r' , shape=(wikiaab_passages.num_rows, 1_28) , )
snake_case: List[str] = faiss.IndexFlatIP(1_28 )
snake_case: List[Any] = faiss.index_cpu_to_gpu(__A , 1 , __A )
wikiaab_gpu_index_flat.add(__A ) # TODO fix for larger GPU
else:
snake_case , snake_case: Optional[Any] = (None, None)
snake_case: List[Any] = Elasticsearch([{'host': 'localhost', 'port': '9200'}] )
return (wikiaab_passages, wikiaab_gpu_index_flat, es_client)
@st.cache(allow_output_mutation=__A )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: str = datasets.load_dataset('eli5' , name='LFQA_reddit' )
snake_case: Optional[int] = elia['train_eli5']
snake_case: Optional[Any] = np.memmap(
'eli5_questions_reps.dat' , dtype='float32' , mode='r' , shape=(elia_train.num_rows, 1_28) )
snake_case: List[str] = faiss.IndexFlatIP(1_28 )
eli5_train_q_index.add(__A )
return (elia_train, eli5_train_q_index)
__UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase = load_indexes()
__UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase = load_models()
__UpperCAmelCase , __UpperCAmelCase = load_train_data()
def lowerCAmelCase_ ( __A : Union[str, Any] , __A : List[Any]=10 ):
'''simple docstring'''
snake_case: List[Any] = embed_questions_for_retrieval([question] , __A , __A )
snake_case , snake_case: Optional[int] = eli5_train_q_index.search(__A , __A )
snake_case: Optional[int] = [elia_train[int(__A )] for i in I[0]]
return nn_examples
def lowerCAmelCase_ ( __A : List[str] , __A : Dict="wiki40b" , __A : Optional[Any]="dense" , __A : Union[str, Any]=10 ):
'''simple docstring'''
if source == "none":
snake_case , snake_case: List[str] = (' <P> '.join(['' for _ in range(11 )] ).strip(), [])
else:
if method == "dense":
snake_case , snake_case: Optional[Any] = query_qa_dense_index(
__A , __A , __A , __A , __A , __A )
else:
snake_case , snake_case: Any = query_es_index(
__A , __A , index_name='english_wiki40b_snippets_100w' , n_results=__A , )
snake_case: Union[str, Any] = [
(res['article_title'], res['section_title'].strip(), res['score'], res['passage_text']) for res in hit_lst
]
snake_case: Dict = 'question: {} context: {}'.format(__A , __A )
return question_doc, support_list
@st.cache(
hash_funcs={
torch.Tensor: (lambda __A : None),
transformers.models.bart.tokenization_bart.BartTokenizer: (lambda __A : None),
} )
def lowerCAmelCase_ ( __A : Tuple , __A : Optional[int] , __A : str , __A : List[str]=64 , __A : Any=2_56 , __A : Dict=False , __A : Optional[Any]=2 , __A : List[str]=0.95 , __A : Any=0.8 ):
'''simple docstring'''
with torch.no_grad():
snake_case: Optional[Any] = qa_sas_generate(
__A , __A , __A , num_answers=1 , num_beams=__A , min_len=__A , max_len=__A , do_sample=__A , temp=__A , top_p=__A , top_k=__A , max_input_length=10_24 , device='cuda:0' , )[0]
return (answer, support_list)
st.title("Long Form Question Answering with ELI5")
# Start sidebar
__UpperCAmelCase = "<img src='https://huggingface.co/front/assets/huggingface_logo.svg'>"
__UpperCAmelCase = "\n<html>\n <head>\n <style>\n .img-container {\n padding-left: 90px;\n padding-right: 90px;\n padding-top: 50px;\n padding-bottom: 50px;\n background-color: #f0f3f9;\n }\n </style>\n </head>\n <body>\n <span class=\"img-container\"> <!-- Inline parent element -->\n %s\n </span>\n </body>\n</html>\n" % (
header_html,
)
st.sidebar.markdown(
header_full,
unsafe_allow_html=True,
)
# Long Form QA with ELI5 and Wikipedia
__UpperCAmelCase = "\nThis demo presents a model trained to [provide long-form answers to open-domain questions](https://yjernite.github.io/lfqa.html).\nFirst, a document retriever fetches a set of relevant Wikipedia passages given the question from the [Wiki40b](https://research.google/pubs/pub49029/) dataset,\na pre-processed fixed snapshot of Wikipedia.\n"
st.sidebar.markdown(description, unsafe_allow_html=True)
__UpperCAmelCase = [
"Answer the question",
"View the retrieved document only",
"View the most similar ELI5 question and answer",
"Show me everything, please!",
]
__UpperCAmelCase = st.sidebar.checkbox("Demo options")
if demo_options:
__UpperCAmelCase = st.sidebar.selectbox(
"",
action_list,
index=3,
)
__UpperCAmelCase = action_list.index(action_st)
__UpperCAmelCase = st.sidebar.selectbox(
"",
["Show full text of passages", "Show passage section titles"],
index=0,
)
__UpperCAmelCase = show_type == "Show full text of passages"
else:
__UpperCAmelCase = 3
__UpperCAmelCase = True
__UpperCAmelCase = st.sidebar.checkbox("Retrieval options")
if retrieval_options:
__UpperCAmelCase = "\n ### Information retriever options\n\n The **sparse** retriever uses ElasticSearch, while the **dense** retriever uses max-inner-product search between a question and passage embedding\n trained using the [ELI5](https://arxiv.org/abs/1907.09190) questions-answer pairs.\n The answer is then generated by sequence to sequence model which takes the question and retrieved document as input.\n "
st.sidebar.markdown(retriever_info)
__UpperCAmelCase = st.sidebar.selectbox("Which Wikipedia format should the model use?", ["wiki40b", "none"])
__UpperCAmelCase = st.sidebar.selectbox("Which Wikipedia indexer should the model use?", ["dense", "sparse", "mixed"])
else:
__UpperCAmelCase = "wiki40b"
__UpperCAmelCase = "dense"
__UpperCAmelCase = "beam"
__UpperCAmelCase = 2
__UpperCAmelCase = 64
__UpperCAmelCase = 256
__UpperCAmelCase = None
__UpperCAmelCase = None
__UpperCAmelCase = st.sidebar.checkbox("Generation options")
if generate_options:
__UpperCAmelCase = "\n ### Answer generation options\n\n The sequence-to-sequence model was initialized with [BART](https://huggingface.co/facebook/bart-large)\n weights and fine-tuned on the ELI5 QA pairs and retrieved documents. You can use the model for greedy decoding with\n **beam** search, or **sample** from the decoder's output probabilities.\n "
st.sidebar.markdown(generate_info)
__UpperCAmelCase = st.sidebar.selectbox("Would you like to use beam search or sample an answer?", ["beam", "sampled"])
__UpperCAmelCase = st.sidebar.slider(
"Minimum generation length", min_value=8, max_value=256, value=64, step=8, format=None, key=None
)
__UpperCAmelCase = st.sidebar.slider(
"Maximum generation length", min_value=64, max_value=512, value=256, step=16, format=None, key=None
)
if sampled == "beam":
__UpperCAmelCase = st.sidebar.slider("Beam size", min_value=1, max_value=8, value=2, step=None, format=None, key=None)
else:
__UpperCAmelCase = st.sidebar.slider(
"Nucleus sampling p", min_value=0.1, max_value=1.0, value=0.95, step=0.01, format=None, key=None
)
__UpperCAmelCase = st.sidebar.slider(
"Temperature", min_value=0.1, max_value=1.0, value=0.7, step=0.01, format=None, key=None
)
__UpperCAmelCase = None
# start main text
__UpperCAmelCase = [
"<MY QUESTION>",
"How do people make chocolate?",
"Why do we get a fever when we are sick?",
"How can different animals perceive different colors?",
"What is natural language processing?",
"What's the best way to treat a sunburn?",
"What exactly are vitamins ?",
"How does nuclear energy provide electricity?",
"What's the difference between viruses and bacteria?",
"Why are flutes classified as woodwinds when most of them are made out of metal ?",
"Why do people like drinking coffee even though it tastes so bad?",
"What happens when wine ages? How does it make the wine taste better?",
"If an animal is an herbivore, where does it get the protein that it needs to survive if it only eats grass?",
"How can we set a date to the beginning or end of an artistic period? Doesn't the change happen gradually?",
"How does New Zealand have so many large bird predators?",
]
__UpperCAmelCase = st.selectbox(
"What would you like to ask? ---- select <MY QUESTION> to enter a new query",
questions_list,
index=1,
)
if question_s == "<MY QUESTION>":
__UpperCAmelCase = st.text_input("Enter your question here:", "")
else:
__UpperCAmelCase = question_s
if st.button("Show me!"):
if action in [0, 1, 3]:
if index_type == "mixed":
__UpperCAmelCase , __UpperCAmelCase = make_support(question, source=wiki_source, method="dense", n_results=10)
__UpperCAmelCase , __UpperCAmelCase = make_support(question, source=wiki_source, method="sparse", n_results=10)
__UpperCAmelCase = []
for res_d, res_s in zip(support_list_dense, support_list_sparse):
if tuple(res_d) not in support_list:
support_list += [tuple(res_d)]
if tuple(res_s) not in support_list:
support_list += [tuple(res_s)]
__UpperCAmelCase = support_list[:10]
__UpperCAmelCase = "<P> " + " <P> ".join([res[-1] for res in support_list])
else:
__UpperCAmelCase , __UpperCAmelCase = make_support(question, source=wiki_source, method=index_type, n_results=10)
if action in [0, 3]:
__UpperCAmelCase , __UpperCAmelCase = answer_question(
question_doc,
sas_model,
sas_tokenizer,
min_len=min_len,
max_len=int(max_len),
sampling=(sampled == "sampled"),
n_beams=n_beams,
top_p=top_p,
temp=temp,
)
st.markdown("### The model generated answer is:")
st.write(answer)
if action in [0, 1, 3] and wiki_source != "none":
st.markdown("--- \n ### The model is drawing information from the following Wikipedia passages:")
for i, res in enumerate(support_list):
__UpperCAmelCase = "https://en.wikipedia.org/wiki/{}".format(res[0].replace(" ", "_"))
__UpperCAmelCase = res[1].strip()
if sec_titles == "":
__UpperCAmelCase = "[{}]({})".format(res[0], wiki_url)
else:
__UpperCAmelCase = sec_titles.split(" & ")
__UpperCAmelCase = " & ".join(
["[{}]({}#{})".format(sec.strip(), wiki_url, sec.strip().replace(" ", "_")) for sec in sec_list]
)
st.markdown(
"{0:02d} - **Article**: {1:<18} <br> _Section_: {2}".format(i + 1, res[0], sections),
unsafe_allow_html=True,
)
if show_passages:
st.write(
"> <span style=\"font-family:arial; font-size:10pt;\">" + res[-1] + "</span>", unsafe_allow_html=True
)
if action in [2, 3]:
__UpperCAmelCase = find_nearest_training(question)
__UpperCAmelCase = nn_train_list[0]
st.markdown(
"--- \n ### The most similar question in the ELI5 training set was: \n\n {}".format(train_exple["title"])
)
__UpperCAmelCase = [
"{}. {}".format(i + 1, " \n".join([line.strip() for line in ans.split("\n") if line.strip() != ""]))
for i, (ans, sc) in enumerate(zip(train_exple["answers"]["text"], train_exple["answers"]["score"]))
if i == 0 or sc > 2
]
st.markdown("##### Its answers were: \n\n {}".format("\n".join(answers_st)))
__UpperCAmelCase = "\n---\n\n**Disclaimer**\n\n*The intent of this app is to provide some (hopefully entertaining) insights into the behavior of a current LFQA system.\nEvaluating biases of such a model and ensuring factual generations are still very much open research problems.\nTherefore, until some significant progress is achieved, we caution against using the generated answers for practical purposes.*\n"
st.sidebar.markdown(disclaimer, unsafe_allow_html=True) | 692 |
'''simple docstring'''
from operator import delitem, getitem, setitem
import pytest
from data_structures.hashing.hash_map import HashMap
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
return getitem, k
def lowerCAmelCase_ ( __A : Any , __A : Optional[int] ):
'''simple docstring'''
return setitem, k, v
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
return delitem, k
def lowerCAmelCase_ ( __A : str , __A : int , *__A : Tuple ):
'''simple docstring'''
try:
return fun(__A , *__A ), None
except Exception as e:
return None, e
__UpperCAmelCase = (
_set("key_a", "val_a"),
_set("key_b", "val_b"),
)
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_a", "val_b"),
]
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_b", "val_b"),
_del("key_a"),
_del("key_b"),
_set("key_a", "val_a"),
_del("key_a"),
]
__UpperCAmelCase = [
_get("key_a"),
_del("key_a"),
_set("key_a", "val_a"),
_del("key_a"),
_del("key_a"),
_get("key_a"),
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
*[_del(x) for x in range(5)],
_set("key_a", "val_b"),
]
@pytest.mark.parametrize(
'operations' , (
pytest.param(_add_items , id='add items' ),
pytest.param(_overwrite_items , id='overwrite items' ),
pytest.param(_delete_items , id='delete items' ),
pytest.param(_access_absent_items , id='access absent items' ),
pytest.param(_add_with_resize_up , id='add with resize up' ),
pytest.param(_add_with_resize_down , id='add with resize down' ),
) , )
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[Any] = HashMap(initial_block_size=4 )
snake_case: List[Any] = {}
for _, (fun, *args) in enumerate(__A ):
snake_case , snake_case: Optional[int] = _run_operation(__A , __A , *__A )
snake_case , snake_case: str = _run_operation(__A , __A , *__A )
assert my_res == py_res
assert str(__A ) == str(__A )
assert set(__A ) == set(__A )
assert len(__A ) == len(__A )
assert set(my.items() ) == set(py.items() )
def lowerCAmelCase_ ( ):
'''simple docstring'''
def is_public(__A : str ) -> bool:
return not name.startswith('_' )
snake_case: Dict = {name for name in dir({} ) if is_public(__A )}
snake_case: List[str] = {name for name in dir(HashMap() ) if is_public(__A )}
assert dict_public_names > hash_public_names | 692 | 1 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : str , __A : list[str] ):
'''simple docstring'''
snake_case: Optional[int] = ''
for word_or_phrase in separated:
if not isinstance(__A , __A ):
raise Exception('join() accepts only strings to be joined' )
joined += word_or_phrase + separator
return joined.strip(__A )
if __name__ == "__main__":
from doctest import testmod
testmod() | 692 |
'''simple docstring'''
import argparse
import json
import os
import fairseq
import torch
from torch import nn
from transformers import (
SpeechaTextaConfig,
SpeechaTextaForCausalLM,
SpeechaTextaTokenizer,
SpeechEncoderDecoderConfig,
SpeechEncoderDecoderModel,
WavaVecaConfig,
WavaVecaFeatureExtractor,
WavaVecaModel,
logging,
)
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"post_extract_proj": "feature_projection.projection",
"encoder.pos_conv.0": "encoder.pos_conv_embed.conv",
"self_attn.k_proj": "encoder.layers.*.attention.k_proj",
"self_attn.v_proj": "encoder.layers.*.attention.v_proj",
"self_attn.q_proj": "encoder.layers.*.attention.q_proj",
"self_attn.out_proj": "encoder.layers.*.attention.out_proj",
"self_attn_layer_norm": "encoder.layers.*.layer_norm",
"fc1": "encoder.layers.*.feed_forward.intermediate_dense",
"fc2": "encoder.layers.*.feed_forward.output_dense",
"final_layer_norm": "encoder.layers.*.final_layer_norm",
"encoder.layer_norm": "encoder.layer_norm",
"w2v_model.layer_norm": "feature_projection.layer_norm",
"quantizer.weight_proj": "quantizer.weight_proj",
"quantizer.vars": "quantizer.codevectors",
"project_q": "project_q",
"final_proj": "project_hid",
"w2v_encoder.proj": "lm_head",
"mask_emb": "masked_spec_embed",
}
__UpperCAmelCase = [
"lm_head",
"quantizer.weight_proj",
"quantizer.codevectors",
"project_q",
"project_hid",
]
def lowerCAmelCase_ ( __A : Any , __A : Optional[Any] , __A : Union[str, Any] , __A : int , __A : Optional[int] ):
'''simple docstring'''
for attribute in key.split('.' ):
snake_case: List[str] = getattr(__A , __A )
if weight_type is not None:
snake_case: Optional[int] = getattr(__A , __A ).shape
else:
snake_case: Optional[int] = hf_pointer.shape
assert hf_shape == value.shape, (
f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
snake_case: Optional[int] = value
elif weight_type == "weight_g":
snake_case: List[str] = value
elif weight_type == "weight_v":
snake_case: Dict = value
elif weight_type == "bias":
snake_case: Optional[Any] = value
else:
snake_case: int = value
logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def lowerCAmelCase_ ( __A : List[Any] , __A : List[str] ):
'''simple docstring'''
snake_case: List[Any] = []
snake_case: List[Any] = fairseq_model.state_dict()
snake_case: Union[str, Any] = hf_model.feature_extractor
# if encoder has different dim to decoder -> use proj_weight
snake_case: Dict = None
for name, value in fairseq_dict.items():
snake_case: Tuple = False
if "conv_layers" in name:
load_conv_layer(
__A , __A , __A , __A , hf_model.config.feat_extract_norm == 'group' , )
snake_case: List[Any] = True
elif name.split('.' )[0] == "proj":
snake_case: List[Any] = fairseq_model.proj
snake_case: int = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
snake_case: int = True
if "*" in mapped_key:
snake_case: List[str] = name.split(__A )[0].split('.' )[-2]
snake_case: Dict = mapped_key.replace('*' , __A )
if "weight_g" in name:
snake_case: Tuple = 'weight_g'
elif "weight_v" in name:
snake_case: int = 'weight_v'
elif "bias" in name:
snake_case: Tuple = 'bias'
elif "weight" in name:
snake_case: List[Any] = 'weight'
else:
snake_case: Any = None
set_recursively(__A , __A , __A , __A , __A )
continue
if not is_used:
unused_weights.append(__A )
logger.warning(f"""Unused weights: {unused_weights}""" )
return proj_weight
def lowerCAmelCase_ ( __A : List[str] , __A : List[Any] , __A : int , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = full_name.split('conv_layers.' )[-1]
snake_case: Tuple = name.split('.' )
snake_case: Any = int(items[0] )
snake_case: Optional[int] = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
snake_case: Tuple = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
snake_case: int = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
snake_case: Any = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
snake_case: str = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(__A )
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case , snake_case: List[Any] = emb.weight.shape
snake_case: Optional[int] = nn.Linear(__A , __A , bias=__A )
snake_case: Any = emb.weight.data
return lin_layer
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
with open(__A , 'r' , encoding='utf-8' ) as f:
snake_case: List[Any] = f.readlines()
snake_case: Any = [line.split(' ' )[0] for line in lines]
snake_case: int = len(__A )
snake_case: Dict = {
'<s>': 0,
'<pad>': 1,
'</s>': 2,
'<unk>': 3,
}
vocab_dict.update(dict(zip(__A , range(4 , num_words + 4 ) ) ) )
return vocab_dict
@torch.no_grad()
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Dict , __A : Any , __A : List[Any] , __A : int , __A : str , ):
'''simple docstring'''
snake_case: Union[str, Any] = WavaVecaConfig.from_pretrained(__A )
snake_case: str = SpeechaTextaConfig.from_pretrained(
__A , vocab_size=__A , decoder_layers=__A , do_stable_layer_norm=__A )
snake_case: List[str] = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=1_60_00 , padding_value=0 , do_normalize=__A , return_attention_mask=__A , )
snake_case , snake_case , snake_case: List[str] = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
snake_case: List[Any] = model[0].eval()
# set weights for wav2vec2 encoder
snake_case: Optional[Any] = WavaVecaModel(__A )
snake_case: Any = recursively_load_weights_wavaveca(model.encoder , __A )
snake_case: Union[str, Any] = SpeechaTextaForCausalLM(__A )
snake_case , snake_case: Optional[Any] = hf_decoder.model.decoder.load_state_dict(model.decoder.state_dict() , strict=__A )
# set output linear layer
unexpected_keys.remove('embed_out' )
snake_case: str = nn.Parameter(model.decoder.embed_out.detach() )
# layer norm is init to identity matrix so leaving it is fine
logger.warning(f"""The following keys are missing when loading the decoder weights: {missing_keys}""" )
logger.warning(f"""The following keys are unexpected when loading the decoder weights: {unexpected_keys}""" )
snake_case: int = SpeechEncoderDecoderModel(encoder=__A , decoder=__A )
snake_case: List[Any] = False
# add projection layer
snake_case: Union[str, Any] = nn.Parameter(projection_layer.weight )
snake_case: Union[str, Any] = nn.Parameter(projection_layer.bias )
snake_case: List[Any] = create_vocab_dict(__A )
with open(os.path.join(__A , 'vocab.json' ) , 'w' ) as fp:
json.dump(__A , __A )
snake_case: Union[str, Any] = SpeechaTextaTokenizer(os.path.join(__A , 'vocab.json' ) )
tokenizer.save_pretrained(__A )
snake_case: Tuple = hf_wavavec.config.to_dict()
snake_case: int = tokenizer.pad_token_id
snake_case: Dict = tokenizer.bos_token_id
snake_case: Optional[int] = tokenizer.eos_token_id
snake_case: Dict = 'speech_to_text_2'
snake_case: Optional[Any] = 'wav2vec2'
snake_case: Tuple = SpeechEncoderDecoderConfig.from_dict(__A )
hf_wavavec.save_pretrained(__A )
feature_extractor.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
parser.add_argument("--dict_path", default=None, type=str, help="Path to dict of fine-tuned model")
parser.add_argument(
"--encoder_config_path",
default="facebook/wav2vec2-large-lv60",
type=str,
help="Path to hf encoder wav2vec2 checkpoint config",
)
parser.add_argument(
"--decoder_config_path",
default="facebook/s2t-small-mustc-en-fr-st",
type=str,
help="Path to hf decoder s2t checkpoint config",
)
parser.add_argument("--vocab_size", default=10_224, type=int, help="Vocab size of decoder")
parser.add_argument("--num_decoder_layers", default=7, type=int, help="Number of decoder layers")
__UpperCAmelCase = parser.parse_args()
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.dict_path,
encoder_config_path=args.encoder_config_path,
decoder_config_path=args.decoder_config_path,
vocab_size=args.vocab_size,
num_decoder_layers=args.num_decoder_layers,
) | 692 | 1 |
'''simple docstring'''
import logging
import os
import sys
from dataclasses import dataclass, field
from typing import Optional
import numpy as np
import torch
from datasets import load_dataset
from torchvision.transforms import Compose, Lambda, Normalize, RandomHorizontalFlip, RandomResizedCrop, ToTensor
import transformers
from transformers import (
CONFIG_MAPPING,
IMAGE_PROCESSOR_MAPPING,
MODEL_FOR_MASKED_IMAGE_MODELING_MAPPING,
AutoConfig,
AutoImageProcessor,
AutoModelForMaskedImageModeling,
HfArgumentParser,
Trainer,
TrainingArguments,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version, send_example_telemetry
from transformers.utils.versions import require_version
__UpperCAmelCase = logging.getLogger(__name__)
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.31.0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt")
__UpperCAmelCase = list(MODEL_FOR_MASKED_IMAGE_MODELING_MAPPING.keys())
__UpperCAmelCase = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default="cifar10" , metadata={"help": "Name of a dataset from the datasets package"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "The column name of the images in the files. If not set, will try to use 'image' or 'img'."} , )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A folder containing the training data."} )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A folder containing the validation data."} )
__UpperCamelCase = field(
default=0.15 , metadata={"help": "Percent to split off of train for validation."} )
__UpperCamelCase = field(default=32 , metadata={"help": "The size of the square patches to use for masking."} )
__UpperCamelCase = field(
default=0.6 , metadata={"help": "Percentage of patches to mask."} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
)
} , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = {}
if self.train_dir is not None:
snake_case: Tuple = self.train_dir
if self.validation_dir is not None:
snake_case: str = self.validation_dir
snake_case: int = data_files if data_files else None
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"The model checkpoint for weights initialization. Can be a local path to a pytorch_model.bin or a "
"checkpoint identifier on the hub. "
"Don't set if you want to train a model from scratch."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "If training from scratch, pass a model type from the list: " + ", ".join(snake_case )} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Override some existing default config settings when a model is trained from scratch. Example: "
"n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index"
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Where do you want to store (cache) the pretrained models/datasets downloaded from the hub"} , )
__UpperCamelCase = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
__UpperCamelCase = field(default=snake_case , metadata={"help": "Name or path of preprocessor config."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"The size (resolution) of each image. If not specified, will use `image_size` of the configuration."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"The size (resolution) of each patch. If not specified, will use `patch_size` of the configuration."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Stride to use for the encoder."} , )
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__=1_92 , SCREAMING_SNAKE_CASE__=32 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=0.6 ):
'''simple docstring'''
snake_case: str = input_size
snake_case: Dict = mask_patch_size
snake_case: List[str] = model_patch_size
snake_case: List[Any] = mask_ratio
if self.input_size % self.mask_patch_size != 0:
raise ValueError('Input size must be divisible by mask patch size' )
if self.mask_patch_size % self.model_patch_size != 0:
raise ValueError('Mask patch size must be divisible by model patch size' )
snake_case: Union[str, Any] = self.input_size // self.mask_patch_size
snake_case: Dict = self.mask_patch_size // self.model_patch_size
snake_case: str = self.rand_size**2
snake_case: Optional[int] = int(np.ceil(self.token_count * self.mask_ratio ) )
def __call__( self ):
'''simple docstring'''
snake_case: Tuple = np.random.permutation(self.token_count )[: self.mask_count]
snake_case: List[Any] = np.zeros(self.token_count , dtype=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = 1
snake_case: Optional[int] = mask.reshape((self.rand_size, self.rand_size) )
snake_case: int = mask.repeat(self.scale , axis=0 ).repeat(self.scale , axis=1 )
return torch.tensor(mask.flatten() )
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
snake_case: Union[str, Any] = torch.stack([example['pixel_values'] for example in examples] )
snake_case: List[str] = torch.stack([example['mask'] for example in examples] )
return {"pixel_values": pixel_values, "bool_masked_pos": mask}
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: int = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case , snake_case , snake_case: Tuple = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case , snake_case , snake_case: Any = parser.parse_args_into_dataclasses()
# Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The
# information sent is the one passed as arguments along with your Python/PyTorch versions.
send_example_telemetry('run_mim' , __A , __A )
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
if training_args.should_log:
# The default of training_args.log_level is passive, so we set log level at info here to have that default.
transformers.utils.logging.set_verbosity_info()
snake_case: Optional[int] = training_args.get_process_log_level()
logger.setLevel(__A )
transformers.utils.logging.set_verbosity(__A )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"""
+ f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" )
logger.info(f"""Training/evaluation parameters {training_args}""" )
# Detecting last checkpoint.
snake_case: Optional[Any] = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case: int = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"""Output directory ({training_args.output_dir}) already exists and is not empty. """
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Initialize our dataset.
snake_case: List[str] = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , data_files=data_args.data_files , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
# If we don't have a validation split, split off a percentage of train as validation.
snake_case: int = None if 'validation' in ds.keys() else data_args.train_val_split
if isinstance(data_args.train_val_split , __A ) and data_args.train_val_split > 0.0:
snake_case: List[Any] = ds['train'].train_test_split(data_args.train_val_split )
snake_case: str = split['train']
snake_case: Optional[int] = split['test']
# Create config
# Distributed training:
# The .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case: Optional[int] = {
'cache_dir': model_args.cache_dir,
'revision': model_args.model_revision,
'use_auth_token': True if model_args.use_auth_token else None,
}
if model_args.config_name_or_path:
snake_case: List[str] = AutoConfig.from_pretrained(model_args.config_name_or_path , **__A )
elif model_args.model_name_or_path:
snake_case: int = AutoConfig.from_pretrained(model_args.model_name_or_path , **__A )
else:
snake_case: Any = CONFIG_MAPPING[model_args.model_type]()
logger.warning('You are instantiating a new config instance from scratch.' )
if model_args.config_overrides is not None:
logger.info(f"""Overriding config: {model_args.config_overrides}""" )
config.update_from_string(model_args.config_overrides )
logger.info(f"""New config: {config}""" )
# make sure the decoder_type is "simmim" (only relevant for BEiT)
if hasattr(__A , 'decoder_type' ):
snake_case: int = 'simmim'
# adapt config
snake_case: int = model_args.image_size if model_args.image_size is not None else config.image_size
snake_case: Union[str, Any] = model_args.patch_size if model_args.patch_size is not None else config.patch_size
snake_case: Optional[Any] = (
model_args.encoder_stride if model_args.encoder_stride is not None else config.encoder_stride
)
config.update(
{
'image_size': model_args.image_size,
'patch_size': model_args.patch_size,
'encoder_stride': model_args.encoder_stride,
} )
# create image processor
if model_args.image_processor_name:
snake_case: Union[str, Any] = AutoImageProcessor.from_pretrained(model_args.image_processor_name , **__A )
elif model_args.model_name_or_path:
snake_case: str = AutoImageProcessor.from_pretrained(model_args.model_name_or_path , **__A )
else:
snake_case: Dict = {
conf.model_type: image_processor_class for conf, image_processor_class in IMAGE_PROCESSOR_MAPPING.items()
}
snake_case: Optional[int] = IMAGE_PROCESSOR_TYPES[model_args.model_type]()
# create model
if model_args.model_name_or_path:
snake_case: Optional[Any] = AutoModelForMaskedImageModeling.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
else:
logger.info('Training new model from scratch' )
snake_case: int = AutoModelForMaskedImageModeling.from_config(__A )
if training_args.do_train:
snake_case: Any = ds['train'].column_names
else:
snake_case: str = ds['validation'].column_names
if data_args.image_column_name is not None:
snake_case: Tuple = data_args.image_column_name
elif "image" in column_names:
snake_case: List[Any] = 'image'
elif "img" in column_names:
snake_case: Optional[int] = 'img'
else:
snake_case: Optional[Any] = column_names[0]
# transformations as done in original SimMIM paper
# source: https://github.com/microsoft/SimMIM/blob/main/data/data_simmim.py
snake_case: List[Any] = Compose(
[
Lambda(lambda __A : img.convert('RGB' ) if img.mode != "RGB" else img ),
RandomResizedCrop(model_args.image_size , scale=(0.67, 1.0) , ratio=(3.0 / 4.0, 4.0 / 3.0) ),
RandomHorizontalFlip(),
ToTensor(),
Normalize(mean=image_processor.image_mean , std=image_processor.image_std ),
] )
# create mask generator
snake_case: List[Any] = MaskGenerator(
input_size=model_args.image_size , mask_patch_size=data_args.mask_patch_size , model_patch_size=model_args.patch_size , mask_ratio=data_args.mask_ratio , )
def preprocess_images(__A : str ):
snake_case: Optional[Any] = [transforms(__A ) for image in examples[image_column_name]]
snake_case: Optional[Any] = [mask_generator() for i in range(len(examples[image_column_name] ) )]
return examples
if training_args.do_train:
if "train" not in ds:
raise ValueError('--do_train requires a train dataset' )
if data_args.max_train_samples is not None:
snake_case: Any = ds['train'].shuffle(seed=training_args.seed ).select(range(data_args.max_train_samples ) )
# Set the training transforms
ds["train"].set_transform(__A )
if training_args.do_eval:
if "validation" not in ds:
raise ValueError('--do_eval requires a validation dataset' )
if data_args.max_eval_samples is not None:
snake_case: Optional[Any] = (
ds['validation'].shuffle(seed=training_args.seed ).select(range(data_args.max_eval_samples ) )
)
# Set the validation transforms
ds["validation"].set_transform(__A )
# Initialize our trainer
snake_case: Optional[int] = Trainer(
model=__A , args=__A , train_dataset=ds['train'] if training_args.do_train else None , eval_dataset=ds['validation'] if training_args.do_eval else None , tokenizer=__A , data_collator=__A , )
# Training
if training_args.do_train:
snake_case: Dict = None
if training_args.resume_from_checkpoint is not None:
snake_case: Union[str, Any] = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
snake_case: Optional[Any] = last_checkpoint
snake_case: int = trainer.train(resume_from_checkpoint=__A )
trainer.save_model()
trainer.log_metrics('train' , train_result.metrics )
trainer.save_metrics('train' , train_result.metrics )
trainer.save_state()
# Evaluation
if training_args.do_eval:
snake_case: Dict = trainer.evaluate()
trainer.log_metrics('eval' , __A )
trainer.save_metrics('eval' , __A )
# Write model card and (optionally) push to hub
snake_case: str = {
'finetuned_from': model_args.model_name_or_path,
'tasks': 'masked-image-modeling',
'dataset': data_args.dataset_name,
'tags': ['masked-image-modeling'],
}
if training_args.push_to_hub:
trainer.push_to_hub(**__A )
else:
trainer.create_model_card(**__A )
if __name__ == "__main__":
main() | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : int = 1_00 ):
'''simple docstring'''
snake_case: List[str] = n * (n + 1) * (2 * n + 1) / 6
snake_case: List[Any] = (n * (n + 1) / 2) ** 2
return int(square_of_sum - sum_of_squares )
if __name__ == "__main__":
print(F'{solution() = }') | 692 | 1 |
'''simple docstring'''
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"sail/poolformer_s12": "https://huggingface.co/sail/poolformer_s12/resolve/main/config.json",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "poolformer"
def __init__( self , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=16 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[64, 1_28, 3_20, 5_12] , SCREAMING_SNAKE_CASE__=[7, 3, 3, 3] , SCREAMING_SNAKE_CASE__=[4, 2, 2, 2] , SCREAMING_SNAKE_CASE__=[2, 1, 1, 1] , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=0.02 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = num_channels
snake_case: Optional[Any] = patch_size
snake_case: Union[str, Any] = stride
snake_case: List[str] = padding
snake_case: List[Any] = pool_size
snake_case: str = hidden_sizes
snake_case: int = mlp_ratio
snake_case: Optional[Any] = depths
snake_case: Tuple = patch_sizes
snake_case: Dict = strides
snake_case: List[str] = num_encoder_blocks
snake_case: List[Any] = drop_path_rate
snake_case: int = hidden_act
snake_case: List[Any] = use_layer_scale
snake_case: Any = layer_scale_init_value
snake_case: Dict = initializer_range
super().__init__(**SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = version.parse("1.11" )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return 2E-3 | 692 |
'''simple docstring'''
import argparse
from typing import Dict
import tensorflow as tf
import torch
from tqdm import tqdm
from transformers import BigBirdPegasusConfig, BigBirdPegasusForConditionalGeneration
__UpperCAmelCase = [
# tf -> hf
("/", "."),
("layer_", "layers."),
("kernel", "weight"),
("beta", "bias"),
("gamma", "weight"),
("pegasus", "model"),
]
__UpperCAmelCase = [
(".output.dense", ".fc2"),
("intermediate.LayerNorm", "final_layer_norm"),
("intermediate.dense", "fc1"),
]
__UpperCAmelCase = (
INIT_COMMON
+ [
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.out_proj"),
("attention.self", "self_attn"),
("attention.encdec.LayerNorm", "encoder_attn_layer_norm"),
("attention.encdec_output.dense", "encoder_attn.out_proj"),
("attention.encdec", "encoder_attn"),
("key", "k_proj"),
("value", "v_proj"),
("query", "q_proj"),
("decoder.LayerNorm", "decoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = (
INIT_COMMON
+ [
("embeddings.word_embeddings", "shared.weight"),
("embeddings.position_embeddings", "embed_positions.weight"),
("attention.self.LayerNorm", "self_attn_layer_norm"),
("attention.output.dense", "self_attn.output"),
("attention.self", "self_attn.self"),
("encoder.LayerNorm", "encoder.layernorm_embedding"),
]
+ END_COMMON
)
__UpperCAmelCase = [
"encdec/key/bias",
"encdec/query/bias",
"encdec/value/bias",
"self/key/bias",
"self/query/bias",
"self/value/bias",
"encdec_output/dense/bias",
"attention/output/dense/bias",
]
def lowerCAmelCase_ ( __A : Dict , __A : List[Any] ):
'''simple docstring'''
for tf_name, hf_name in patterns:
snake_case: List[Any] = k.replace(__A , __A )
return k
def lowerCAmelCase_ ( __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[int] = BigBirdPegasusConfig(**__A )
snake_case: List[Any] = BigBirdPegasusForConditionalGeneration(__A )
snake_case: Any = torch_model.state_dict()
snake_case: Any = {}
# separating decoder weights
snake_case: Optional[Any] = {k: tf_weights[k] for k in tf_weights if k.startswith('pegasus/decoder' )}
snake_case: Any = {k: tf_weights[k] for k in tf_weights if not k.startswith('pegasus/decoder' )}
for k, v in tqdm(decoder_weights.items() , 'tf -> hf conversion' ):
snake_case: List[str] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Any = DECODER_PATTERNS
snake_case: int = rename_state_dict_key(__A , __A )
if new_k not in state_dict:
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: Optional[Any] = v.T
snake_case: Any = torch.from_numpy(__A )
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
for k, v in tqdm(remaining_weights.items() , 'tf -> hf conversion' ):
snake_case: List[Any] = [k.endswith(__A ) for ending in KEYS_TO_IGNORE]
if any(__A ):
continue
snake_case: Union[str, Any] = REMAINING_PATTERNS
snake_case: str = rename_state_dict_key(__A , __A )
if new_k not in state_dict and k != "pegasus/embeddings/position_embeddings":
raise ValueError(f"""could not find new key {new_k} in state dict. (converted from {k})""" )
if any(True if i in k else False for i in ['dense', 'query', 'key', 'value'] ):
snake_case: int = v.T
snake_case: Any = torch.from_numpy(__A )
if k != "pegasus/embeddings/position_embeddings":
assert v.shape == state_dict[new_k].shape, f"""{new_k}, {k}, {v.shape}, {state_dict[new_k].shape}"""
snake_case: str = mapping['model.embed_positions.weight']
snake_case: Any = mapping.pop('model.embed_positions.weight' )
snake_case , snake_case: Union[str, Any] = torch_model.load_state_dict(__A , strict=__A )
snake_case: Optional[int] = [
k
for k in missing
if k
not in [
'final_logits_bias',
'model.encoder.embed_tokens.weight',
'model.decoder.embed_tokens.weight',
'lm_head.weight',
]
]
assert unexpected_missing == [], f"""no matches found for the following torch keys {unexpected_missing}"""
assert extra == [], f"""no matches found for the following tf keys {extra}"""
return torch_model
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
snake_case: Tuple = tf.train.list_variables(__A )
snake_case: str = {}
snake_case: List[str] = ['global_step']
for name, shape in tqdm(__A , desc='converting tf checkpoint to dict' ):
snake_case: str = any(pat in name for pat in ignore_name )
if skip_key:
continue
snake_case: Any = tf.train.load_variable(__A , __A )
snake_case: Optional[int] = array
return tf_weights
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict ):
'''simple docstring'''
snake_case: int = get_tf_weights_as_numpy(__A )
snake_case: int = convert_bigbird_pegasus(__A , __A )
torch_model.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
parser.add_argument("--tf_ckpt_path", type=str, help="passed to tf.train.list_variables")
parser.add_argument("--save_dir", default=None, type=str, help="Path to the output PyTorch model.")
__UpperCAmelCase = parser.parse_args()
__UpperCAmelCase = {}
convert_bigbird_pegasus_ckpt_to_pytorch(args.tf_ckpt_path, args.save_dir, config_update=config_update) | 692 | 1 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_vision_encoder_decoder": ["VisionEncoderDecoderConfig", "VisionEncoderDecoderOnnxConfig"]
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["VisionEncoderDecoderModel"]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["TFVisionEncoderDecoderModel"]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["FlaxVisionEncoderDecoderModel"]
if TYPE_CHECKING:
from .configuration_vision_encoder_decoder import VisionEncoderDecoderConfig, VisionEncoderDecoderOnnxConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_vision_encoder_decoder import VisionEncoderDecoderModel
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_vision_encoder_decoder import TFVisionEncoderDecoderModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_vision_encoder_decoder import FlaxVisionEncoderDecoderModel
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
snake_case: str = [0] * len(__A )
snake_case: Tuple = []
snake_case: Tuple = [1] * len(__A )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(__A ) ):
if indegree[i] == 0:
queue.append(__A )
while queue:
snake_case: int = queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
snake_case: Any = long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(__A )
print(max(__A ) )
# Adjacency list of Graph
__UpperCAmelCase = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph) | 692 | 1 |
'''simple docstring'''
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = ["image_processor", "tokenizer"]
__UpperCamelCase = "BlipImageProcessor"
__UpperCamelCase = "AutoTokenizer"
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[Any] = False
super().__init__(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.image_processor
def __call__( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
if images is None and text is None:
raise ValueError('You have to specify either images or text.' )
# Get only text
if images is None:
snake_case: List[Any] = self.tokenizer
snake_case: List[str] = self.tokenizer(
text=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , pad_to_multiple_of=SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_overflowing_tokens=SCREAMING_SNAKE_CASE__ , return_special_tokens_mask=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_length=SCREAMING_SNAKE_CASE__ , verbose=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
return text_encoding
# add pixel_values
snake_case: int = self.image_processor(SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
if text is not None:
snake_case: Tuple = self.tokenizer(
text=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , pad_to_multiple_of=SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_overflowing_tokens=SCREAMING_SNAKE_CASE__ , return_special_tokens_mask=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_length=SCREAMING_SNAKE_CASE__ , verbose=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
else:
snake_case: Any = None
if text_encoding is not None:
encoding_image_processor.update(SCREAMING_SNAKE_CASE__ )
return encoding_image_processor
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer.batch_decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer.decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
@property
# Copied from transformers.models.blip.processing_blip.BlipProcessor.model_input_names
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.tokenizer.model_input_names
snake_case: Union[str, Any] = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names ) ) | 692 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
import numpy as np
import pytest
from transformers import BertTokenizer, BertTokenizerFast
from transformers.models.bert.tokenization_bert import VOCAB_FILES_NAMES
from transformers.testing_utils import require_vision
from transformers.utils import FEATURE_EXTRACTOR_NAME, is_vision_available
if is_vision_available():
from PIL import Image
from transformers import ChineseCLIPImageProcessor, ChineseCLIPProcessor
@require_vision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = tempfile.mkdtemp()
snake_case: Optional[Any] = [
'[UNK]',
'[CLS]',
'[SEP]',
'[PAD]',
'[MASK]',
'的',
'价',
'格',
'是',
'15',
'便',
'alex',
'##andra',
'，',
'。',
'-',
't',
'shirt',
]
snake_case: Optional[Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
snake_case: Optional[int] = {
'do_resize': True,
'size': {'height': 2_24, 'width': 2_24},
'do_center_crop': True,
'crop_size': {'height': 18, 'width': 18},
'do_normalize': True,
'image_mean': [0.48_14_54_66, 0.4_57_82_75, 0.40_82_10_73],
'image_std': [0.26_86_29_54, 0.26_13_02_58, 0.27_57_77_11],
'do_convert_rgb': True,
}
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , SCREAMING_SNAKE_CASE__ )
with open(self.image_processor_file , 'w' , encoding='utf-8' ) as fp:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizer.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return BertTokenizerFast.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return ChineseCLIPImageProcessor.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Tuple = [Image.fromarray(np.moveaxis(SCREAMING_SNAKE_CASE__ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_tokenizer()
snake_case: Union[str, Any] = self.get_rust_tokenizer()
snake_case: Union[str, Any] = self.get_image_processor()
snake_case: List[str] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_slow.save_pretrained(self.tmpdirname )
snake_case: List[str] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname , use_fast=SCREAMING_SNAKE_CASE__ )
snake_case: Any = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
processor_fast.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = ChineseCLIPProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor_slow.tokenizer.get_vocab() , tokenizer_slow.get_vocab() )
self.assertEqual(processor_fast.tokenizer.get_vocab() , tokenizer_fast.get_vocab() )
self.assertEqual(tokenizer_slow.get_vocab() , tokenizer_fast.get_vocab() )
self.assertIsInstance(processor_slow.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor_slow.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertEqual(processor_fast.image_processor.to_json_string() , image_processor.to_json_string() )
self.assertIsInstance(processor_slow.image_processor , SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(processor_fast.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=self.get_tokenizer() , image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: Optional[int] = self.get_tokenizer(cls_token='(CLS)' , sep_token='(SEP)' )
snake_case: Union[str, Any] = self.get_image_processor(do_normalize=SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = ChineseCLIPProcessor.from_pretrained(
self.tmpdirname , cls_token='(CLS)' , sep_token='(SEP)' , do_normalize=SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
self.assertIsInstance(processor.tokenizer , SCREAMING_SNAKE_CASE__ )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_image_processor()
snake_case: Tuple = self.get_tokenizer()
snake_case: Optional[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.prepare_image_inputs()
snake_case: List[Any] = image_processor(SCREAMING_SNAKE_CASE__ , return_tensors='np' )
snake_case: Dict = processor(images=SCREAMING_SNAKE_CASE__ , return_tensors='np' )
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_image_processor()
snake_case: Optional[int] = self.get_tokenizer()
snake_case: List[Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Union[str, Any] = processor(text=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer(SCREAMING_SNAKE_CASE__ )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Dict = 'Alexandra，T-shirt的价格是15便士。'
snake_case: Tuple = self.prepare_image_inputs()
snake_case: Any = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , ['input_ids', 'token_type_ids', 'attention_mask', 'pixel_values'] )
# test if it raises when no input is passed
with pytest.raises(SCREAMING_SNAKE_CASE__ ):
processor()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.get_image_processor()
snake_case: str = self.get_tokenizer()
snake_case: Union[str, Any] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = [[1, 4, 5, 8, 1, 0, 8], [3, 4, 3, 1, 1, 8, 9]]
snake_case: int = processor.batch_decode(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = tokenizer.batch_decode(SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_image_processor()
snake_case: Dict = self.get_tokenizer()
snake_case: Optional[int] = ChineseCLIPProcessor(tokenizer=SCREAMING_SNAKE_CASE__ , image_processor=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = 'Alexandra，T-shirt的价格是15便士。'
snake_case: List[Any] = self.prepare_image_inputs()
snake_case: Dict = processor(text=SCREAMING_SNAKE_CASE__ , images=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(list(inputs.keys() ) , processor.model_input_names ) | 692 | 1 |
'''simple docstring'''
import warnings
from ...utils import logging
from .image_processing_chinese_clip import ChineseCLIPImageProcessor
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
warnings.warn(
'The class ChineseCLIPFeatureExtractor is deprecated and will be removed in version 5 of Transformers.'
' Please use ChineseCLIPImageProcessor instead.' , SCREAMING_SNAKE_CASE__ , )
super().__init__(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ) | 692 |
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"microsoft/swinv2-tiny-patch4-window8-256": (
"https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256/resolve/main/config.json"
),
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = "swinv2"
__UpperCamelCase = {
"num_attention_heads": "num_heads",
"num_hidden_layers": "num_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=2_24 , SCREAMING_SNAKE_CASE__=4 , SCREAMING_SNAKE_CASE__=3 , SCREAMING_SNAKE_CASE__=96 , SCREAMING_SNAKE_CASE__=[2, 2, 6, 2] , SCREAMING_SNAKE_CASE__=[3, 6, 12, 24] , SCREAMING_SNAKE_CASE__=7 , SCREAMING_SNAKE_CASE__=4.0 , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=1E-5 , SCREAMING_SNAKE_CASE__=32 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**SCREAMING_SNAKE_CASE__ )
snake_case: int = image_size
snake_case: Union[str, Any] = patch_size
snake_case: List[str] = num_channels
snake_case: Tuple = embed_dim
snake_case: str = depths
snake_case: Any = len(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = num_heads
snake_case: Optional[int] = window_size
snake_case: Any = mlp_ratio
snake_case: Optional[int] = qkv_bias
snake_case: Union[str, Any] = hidden_dropout_prob
snake_case: List[str] = attention_probs_dropout_prob
snake_case: Dict = drop_path_rate
snake_case: List[str] = hidden_act
snake_case: int = use_absolute_embeddings
snake_case: Any = layer_norm_eps
snake_case: Dict = initializer_range
snake_case: List[Any] = encoder_stride
# we set the hidden_size attribute in order to make Swinv2 work with VisionEncoderDecoderModel
# this indicates the channel dimension after the last stage of the model
snake_case: Tuple = int(embed_dim * 2 ** (len(SCREAMING_SNAKE_CASE__ ) - 1) )
snake_case: Union[str, Any] = (0, 0, 0, 0) | 692 | 1 |
'''simple docstring'''
import argparse
import os
from pathlib import Path
import torch
from bark.generation import _load_model as _bark_load_model
from huggingface_hub import hf_hub_download
from transformers import EncodecConfig, EncodecModel, set_seed
from transformers.models.bark.configuration_bark import (
BarkCoarseConfig,
BarkConfig,
BarkFineConfig,
BarkSemanticConfig,
)
from transformers.models.bark.generation_configuration_bark import (
BarkCoarseGenerationConfig,
BarkFineGenerationConfig,
BarkGenerationConfig,
BarkSemanticGenerationConfig,
)
from transformers.models.bark.modeling_bark import BarkCoarseModel, BarkFineModel, BarkModel, BarkSemanticModel
from transformers.utils import logging
logging.set_verbosity_info()
__UpperCAmelCase = logging.get_logger(__name__)
set_seed(770)
__UpperCAmelCase = {
"c_attn": "att_proj",
"c_proj": "out_proj",
"c_fc": "in_proj",
"transformer.": "",
"h.": "layers.",
"ln_1": "layernorm_1",
"ln_2": "layernorm_2",
"ln_f": "layernorm_final",
"wpe": "position_embeds_layer",
"wte": "input_embeds_layer",
}
__UpperCAmelCase = {
"text_small": {
"repo_id": "suno/bark",
"file_name": "text.pt",
},
"coarse_small": {
"repo_id": "suno/bark",
"file_name": "coarse.pt",
},
"fine_small": {
"repo_id": "suno/bark",
"file_name": "fine.pt",
},
"text": {
"repo_id": "suno/bark",
"file_name": "text_2.pt",
},
"coarse": {
"repo_id": "suno/bark",
"file_name": "coarse_2.pt",
},
"fine": {
"repo_id": "suno/bark",
"file_name": "fine_2.pt",
},
}
__UpperCAmelCase = os.path.dirname(os.path.abspath(__file__))
__UpperCAmelCase = os.path.join(os.path.expanduser("~"), ".cache")
__UpperCAmelCase = os.path.join(os.getenv("XDG_CACHE_HOME", default_cache_dir), "suno", "bark_v0")
def lowerCAmelCase_ ( __A : List[str] , __A : Tuple=False ):
'''simple docstring'''
snake_case: str = model_type
if use_small:
key += "_small"
return os.path.join(__A , REMOTE_MODEL_PATHS[key]['file_name'] )
def lowerCAmelCase_ ( __A : Dict , __A : str ):
'''simple docstring'''
os.makedirs(__A , exist_ok=__A )
hf_hub_download(repo_id=__A , filename=__A , local_dir=__A )
def lowerCAmelCase_ ( __A : str , __A : Optional[Any] , __A : Union[str, Any]=False , __A : Union[str, Any]="text" ):
'''simple docstring'''
if model_type == "text":
snake_case: Union[str, Any] = BarkSemanticModel
snake_case: str = BarkSemanticConfig
snake_case: Union[str, Any] = BarkSemanticGenerationConfig
elif model_type == "coarse":
snake_case: Any = BarkCoarseModel
snake_case: Tuple = BarkCoarseConfig
snake_case: List[Any] = BarkCoarseGenerationConfig
elif model_type == "fine":
snake_case: Union[str, Any] = BarkFineModel
snake_case: Union[str, Any] = BarkFineConfig
snake_case: str = BarkFineGenerationConfig
else:
raise NotImplementedError()
snake_case: int = f"""{model_type}_small""" if use_small else model_type
snake_case: List[Any] = REMOTE_MODEL_PATHS[model_key]
if not os.path.exists(__A ):
logger.info(f"""{model_type} model not found, downloading into `{CACHE_DIR}`.""" )
_download(model_info['repo_id'] , model_info['file_name'] )
snake_case: Dict = torch.load(__A , map_location=__A )
# this is a hack
snake_case: Tuple = checkpoint['model_args']
if "input_vocab_size" not in model_args:
snake_case: List[Any] = model_args['vocab_size']
snake_case: int = model_args['vocab_size']
del model_args["vocab_size"]
# convert Bark model arguments to HF Bark model arguments
snake_case: Dict = model_args.pop('n_head' )
snake_case: Tuple = model_args.pop('n_embd' )
snake_case: List[Any] = model_args.pop('n_layer' )
snake_case: Any = ConfigClass(**checkpoint['model_args'] )
snake_case: Tuple = ModelClass(config=__A )
snake_case: int = GenerationConfigClass()
snake_case: List[str] = model_generation_config
snake_case: Optional[Any] = checkpoint['model']
# fixup checkpoint
snake_case: List[Any] = '_orig_mod.'
for k, v in list(state_dict.items() ):
if k.startswith(__A ):
# replace part of the key with corresponding layer name in HF implementation
snake_case: Union[str, Any] = k[len(__A ) :]
for old_layer_name in new_layer_name_dict:
snake_case: Optional[int] = new_k.replace(__A , new_layer_name_dict[old_layer_name] )
snake_case: Optional[int] = state_dict.pop(__A )
snake_case: Union[str, Any] = set(state_dict.keys() ) - set(model.state_dict().keys() )
snake_case: List[Any] = {k for k in extra_keys if not k.endswith('.attn.bias' )}
snake_case: Dict = set(model.state_dict().keys() ) - set(state_dict.keys() )
snake_case: Tuple = {k for k in missing_keys if not k.endswith('.attn.bias' )}
if len(__A ) != 0:
raise ValueError(f"""extra keys found: {extra_keys}""" )
if len(__A ) != 0:
raise ValueError(f"""missing keys: {missing_keys}""" )
model.load_state_dict(__A , strict=__A )
snake_case: int = model.num_parameters(exclude_embeddings=__A )
snake_case: int = checkpoint['best_val_loss'].item()
logger.info(f"""model loaded: {round(n_params/1E6 , 1 )}M params, {round(__A , 3 )} loss""" )
model.eval()
model.to(__A )
del checkpoint, state_dict
return model
def lowerCAmelCase_ ( __A : str , __A : int=False , __A : Union[str, Any]="text" ):
'''simple docstring'''
if model_type not in ("text", "coarse", "fine"):
raise NotImplementedError()
snake_case: Dict = 'cpu' # do conversion on cpu
snake_case: List[Any] = _get_ckpt_path(__A , use_small=__A )
snake_case: Tuple = _load_model(__A , __A , model_type=__A , use_small=__A )
# load bark initial model
snake_case: List[str] = _bark_load_model(__A , 'cpu' , model_type=__A , use_small=__A )
if model_type == "text":
snake_case: str = bark_model['model']
if model.num_parameters(exclude_embeddings=__A ) != bark_model.get_num_params():
raise ValueError('initial and new models don\'t have the same number of parameters' )
# check if same output as the bark model
snake_case: List[Any] = 5
snake_case: str = 10
if model_type in ["text", "coarse"]:
snake_case: List[Any] = torch.randint(2_56 , (batch_size, sequence_length) , dtype=torch.int )
snake_case: List[Any] = bark_model(__A )[0]
snake_case: Optional[Any] = model(__A )
# take last logits
snake_case: Optional[Any] = output_new_model_total.logits[:, [-1], :]
else:
snake_case: Optional[int] = 3
snake_case: str = 8
snake_case: Optional[Any] = torch.randint(2_56 , (batch_size, sequence_length, n_codes_total) , dtype=torch.int )
snake_case: Optional[int] = model(__A , __A )
snake_case: Dict = bark_model(__A , __A )
snake_case: Any = output_new_model_total.logits
# output difference should come from the difference of self-attention implementation design
if output_new_model.shape != output_old_model.shape:
raise ValueError('initial and new outputs don\'t have the same shape' )
if (output_new_model - output_old_model).abs().max().item() > 1E-3:
raise ValueError('initial and new outputs are not equal' )
Path(__A ).mkdir(exist_ok=__A )
model.save_pretrained(__A )
def lowerCAmelCase_ ( __A : Optional[int] , __A : str , __A : int , __A : str , __A : Tuple , __A : Dict , ):
'''simple docstring'''
snake_case: List[Any] = os.path.join(__A , __A )
snake_case: List[Any] = BarkSemanticConfig.from_pretrained(os.path.join(__A , 'config.json' ) )
snake_case: Dict = BarkCoarseConfig.from_pretrained(os.path.join(__A , 'config.json' ) )
snake_case: Tuple = BarkFineConfig.from_pretrained(os.path.join(__A , 'config.json' ) )
snake_case: Tuple = EncodecConfig.from_pretrained('facebook/encodec_24khz' )
snake_case: Tuple = BarkSemanticModel.from_pretrained(__A )
snake_case: int = BarkCoarseModel.from_pretrained(__A )
snake_case: str = BarkFineModel.from_pretrained(__A )
snake_case: Tuple = EncodecModel.from_pretrained('facebook/encodec_24khz' )
snake_case: Optional[Any] = BarkConfig.from_sub_model_configs(
__A , __A , __A , __A )
snake_case: Optional[Any] = BarkGenerationConfig.from_sub_model_configs(
semantic.generation_config , coarseAcoustic.generation_config , fineAcoustic.generation_config )
snake_case: List[Any] = BarkModel(__A )
snake_case: Any = semantic
snake_case: Union[str, Any] = coarseAcoustic
snake_case: Tuple = fineAcoustic
snake_case: Any = codec
snake_case: str = bark_generation_config
Path(__A ).mkdir(exist_ok=__A )
bark.save_pretrained(__A , repo_id=__A , push_to_hub=__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument("model_type", type=str, help="text, coarse or fine.")
parser.add_argument("pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument("--is_small", action="store_true", help="convert the small version instead of the large.")
__UpperCAmelCase = parser.parse_args()
load_model(args.pytorch_dump_folder_path, model_type=args.model_type, use_small=args.is_small) | 692 |
'''simple docstring'''
import os
import sys
import unittest
__UpperCAmelCase = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__))))
sys.path.append(os.path.join(git_repo_path, "utils"))
import check_dummies # noqa: E402
from check_dummies import create_dummy_files, create_dummy_object, find_backend, read_init # noqa: E402
# Align TRANSFORMERS_PATH in check_dummies with the current path
__UpperCAmelCase = os.path.join(git_repo_path, "src", "transformers")
__UpperCAmelCase = "\n{0} = None\n"
__UpperCAmelCase = "\nclass {0}(metaclass=DummyObject):\n _backends = {1}\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, {1})\n"
__UpperCAmelCase = "\ndef {0}(*args, **kwargs):\n requires_backends({0}, {1})\n"
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = find_backend(' _import_structure["models.albert"].append("AlbertTokenizerFast")' )
self.assertIsNone(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = find_backend(' if not is_tokenizers_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tokenizers' )
snake_case: List[Any] = find_backend(' if not is_tensorflow_text_available():' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'tensorflow_text' )
snake_case: int = find_backend(' if not (is_sentencepiece_available() and is_tokenizers_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers' )
snake_case: Optional[Any] = find_backend(
' if not (is_sentencepiece_available() and is_tensorflow_text_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tensorflow_text' )
snake_case: Dict = find_backend(
' if not (is_sentencepiece_available() and is_tokenizers_available() and is_vision_available()):' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'sentencepiece_and_tokenizers_and_vision' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = read_init()
# We don't assert on the exact list of keys to allow for smooth grow of backend-specific objects
self.assertIn('torch' , SCREAMING_SNAKE_CASE__ )
self.assertIn('tensorflow_text' , SCREAMING_SNAKE_CASE__ )
self.assertIn('sentencepiece_and_tokenizers' , SCREAMING_SNAKE_CASE__ )
# Likewise, we can't assert on the exact content of a key
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertModel' , objects['tf'] )
self.assertIn('FlaxBertModel' , objects['flax'] )
self.assertIn('BertModel' , objects['torch'] )
self.assertIn('TFBertTokenizer' , objects['tensorflow_text'] )
self.assertIn('convert_slow_tokenizer' , objects['sentencepiece_and_tokenizers'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = create_dummy_object('CONSTANT' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , '\nCONSTANT = None\n' )
snake_case: Any = create_dummy_object('function' , '\'torch\'' )
self.assertEqual(
SCREAMING_SNAKE_CASE__ , '\ndef function(*args, **kwargs):\n requires_backends(function, \'torch\')\n' )
snake_case: Optional[int] = '\nclass FakeClass(metaclass=DummyObject):\n _backends = \'torch\'\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, \'torch\')\n'
snake_case: Tuple = create_dummy_object('FakeClass' , '\'torch\'' )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = '# This file is autogenerated by the command `make fix-copies`, do not edit.\nfrom ..utils import DummyObject, requires_backends\n\n\nCONSTANT = None\n\n\ndef function(*args, **kwargs):\n requires_backends(function, ["torch"])\n\n\nclass FakeClass(metaclass=DummyObject):\n _backends = ["torch"]\n\n def __init__(self, *args, **kwargs):\n requires_backends(self, ["torch"])\n'
snake_case: Optional[int] = create_dummy_files({'torch': ['CONSTANT', 'function', 'FakeClass']} )
self.assertEqual(dummy_files['torch'] , SCREAMING_SNAKE_CASE__ ) | 692 | 1 |
'''simple docstring'''
import argparse
from torch import nn
# transformers_old should correspond to branch `save_old_prophetnet_model_structure` here
# original prophetnet_checkpoints are saved under `patrickvonplaten/..._old` respectively
from transformers_old.modeling_prophetnet import (
ProphetNetForConditionalGeneration as ProphetNetForConditionalGenerationOld,
)
from transformers_old.modeling_xlm_prophetnet import (
XLMProphetNetForConditionalGeneration as XLMProphetNetForConditionalGenerationOld,
)
from transformers import ProphetNetForConditionalGeneration, XLMProphetNetForConditionalGeneration, logging
__UpperCAmelCase = logging.get_logger(__name__)
logging.set_verbosity_info()
def lowerCAmelCase_ ( __A : str , __A : str ):
'''simple docstring'''
if "xprophetnet" in prophetnet_checkpoint_path:
snake_case: Union[str, Any] = XLMProphetNetForConditionalGenerationOld.from_pretrained(__A )
snake_case , snake_case: Tuple = XLMProphetNetForConditionalGeneration.from_pretrained(
__A , output_loading_info=__A )
else:
snake_case: Optional[int] = ProphetNetForConditionalGenerationOld.from_pretrained(__A )
snake_case , snake_case: List[Any] = ProphetNetForConditionalGeneration.from_pretrained(
__A , output_loading_info=__A )
snake_case: Optional[int] = ['key_proj', 'value_proj', 'query_proj']
snake_case: Optional[Any] = {
'self_attn': 'ngram_self_attn',
'cross_attn': 'encoder_attn',
'cross_attn_layer_norm': 'encoder_attn_layer_norm',
'feed_forward_layer_norm': 'final_layer_norm',
'feed_forward': '',
'intermediate': 'fc1',
'output': 'fc2',
'key_proj': 'k_proj',
'query_proj': 'q_proj',
'value_proj': 'v_proj',
'word_embeddings': 'embed_tokens',
'embeddings_layer_norm': 'emb_layer_norm',
'relative_pos_embeddings': 'relative_linear',
'ngram_embeddings': 'ngram_input_embed',
'position_embeddings': 'embed_positions',
}
for key in loading_info["missing_keys"]:
snake_case: int = key.split('.' )
if attributes[0] == "lm_head":
snake_case: List[str] = prophet
snake_case: Any = prophet_old
else:
snake_case: str = prophet.prophetnet
snake_case: List[str] = prophet_old.model
snake_case: int = False
for attribute in attributes:
if attribute in mapping:
snake_case: Optional[Any] = mapping[attribute]
if not hasattr(__A , __A ) and len(__A ) > 0:
snake_case: List[str] = attribute
elif hasattr(__A , __A ):
snake_case: Any = attribute
if attribute == "weight":
assert old_model.weight.shape == model.weight.shape, "Shapes have to match!"
snake_case: int = old_model.weight
logger.info(f"""{attribute} is initialized.""" )
snake_case: Dict = True
break
elif attribute == "bias":
assert old_model.bias.shape == model.bias.shape, "Shapes have to match!"
snake_case: int = old_model.bias
logger.info(f"""{attribute} is initialized""" )
snake_case: List[str] = True
break
elif attribute in special_keys and hasattr(__A , 'in_proj_weight' ):
snake_case: Tuple = old_model.in_proj_weight.shape[0] // 3
snake_case: Any = getattr(__A , __A )
param.weight.shape == old_model.in_proj_weight[:embed_dim, :].shape, "Shapes have to match"
param.bias.shape == old_model.in_proj_bias[:embed_dim].shape, "Shapes have to match"
if attribute == "query_proj":
snake_case: Optional[Any] = nn.Parameter(old_model.in_proj_weight[:embed_dim, :] )
snake_case: Optional[int] = nn.Parameter(old_model.in_proj_bias[:embed_dim] )
elif attribute == "key_proj":
snake_case: Optional[int] = nn.Parameter(old_model.in_proj_weight[embed_dim : 2 * embed_dim, :] )
snake_case: Union[str, Any] = nn.Parameter(old_model.in_proj_bias[embed_dim : 2 * embed_dim] )
elif attribute == "value_proj":
snake_case: str = nn.Parameter(old_model.in_proj_weight[2 * embed_dim :, :] )
snake_case: Optional[int] = nn.Parameter(old_model.in_proj_bias[2 * embed_dim :] )
snake_case: Optional[int] = True
break
elif attribute == "position_embeddings":
assert (
model.position_embeddings.weight.shape[-1] == old_model.embed_positions.weight.shape[-1]
), "Hidden size has to match"
assert model.position_embeddings.weight.shape[0] == 5_12, "We want 512 position_embeddings."
snake_case: Union[str, Any] = nn.Parameter(old_model.embed_positions.weight[:5_12, :] )
snake_case: Dict = True
break
if attribute.isdigit():
snake_case: Any = model[int(__A )]
snake_case: Optional[Any] = old_model[int(__A )]
else:
snake_case: Dict = getattr(__A , __A )
if old_attribute == "":
snake_case: str = old_model
else:
if not hasattr(__A , __A ):
raise ValueError(f"""{old_model} does not have {old_attribute}""" )
snake_case: Any = getattr(__A , __A )
if not is_key_init:
raise ValueError(f"""{key} was not correctly initialized!""" )
print(f"""Saving model to {pytorch_dump_folder_path}""" )
prophet.save_pretrained(__A )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--prophetnet_checkpoint_path", default=None, type=str, required=True, help="Path the official PyTorch dump."
)
parser.add_argument(
"--pytorch_dump_folder_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
__UpperCAmelCase = parser.parse_args()
convert_prophetnet_checkpoint_to_pytorch(args.prophetnet_checkpoint_path, args.pytorch_dump_folder_path) | 692 |
'''simple docstring'''
import os
import warnings
from typing import List, Optional
from ...tokenization_utils_base import BatchEncoding
from ...utils import logging
from .configuration_rag import RagConfig
__UpperCAmelCase = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = question_encoder
snake_case: Union[str, Any] = generator
snake_case: Optional[int] = self.question_encoder
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if os.path.isfile(SCREAMING_SNAKE_CASE__ ):
raise ValueError(F"""Provided path ({save_directory}) should be a directory, not a file""" )
os.makedirs(SCREAMING_SNAKE_CASE__ , exist_ok=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'question_encoder_tokenizer' )
snake_case: Tuple = os.path.join(SCREAMING_SNAKE_CASE__ , 'generator_tokenizer' )
self.question_encoder.save_pretrained(SCREAMING_SNAKE_CASE__ )
self.generator.save_pretrained(SCREAMING_SNAKE_CASE__ )
@classmethod
def _UpperCamelCase ( cls , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
from ..auto.tokenization_auto import AutoTokenizer
snake_case: int = kwargs.pop('config' , SCREAMING_SNAKE_CASE__ )
if config is None:
snake_case: str = RagConfig.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.question_encoder , subfolder='question_encoder_tokenizer' )
snake_case: Dict = AutoTokenizer.from_pretrained(
SCREAMING_SNAKE_CASE__ , config=config.generator , subfolder='generator_tokenizer' )
return cls(question_encoder=SCREAMING_SNAKE_CASE__ , generator=SCREAMING_SNAKE_CASE__ )
def __call__( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.current_tokenizer(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.batch_decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , *SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.generator.decode(*SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.question_encoder
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.generator
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "longest" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
warnings.warn(
'`prepare_seq2seq_batch` is deprecated and will be removed in version 5 of 🤗 Transformers. Use the '
'regular `__call__` method to prepare your inputs and the tokenizer under the `with_target_tokenizer` '
'context manager to prepare your targets. See the documentation of your specific tokenizer for more '
'details' , SCREAMING_SNAKE_CASE__ , )
if max_length is None:
snake_case: Optional[Any] = self.current_tokenizer.model_max_length
snake_case: int = self(
SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if tgt_texts is None:
return model_inputs
# Process tgt_texts
if max_target_length is None:
snake_case: Any = self.current_tokenizer.model_max_length
snake_case: List[str] = self(
text_target=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , max_length=SCREAMING_SNAKE_CASE__ , truncation=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: Dict = labels['input_ids']
return model_inputs | 692 | 1 |
'''simple docstring'''
import numpy as np
from cva import COLOR_BGR2GRAY, cvtColor, imread
from numpy import array, uinta
from PIL import Image
from digital_image_processing import change_contrast as cc
from digital_image_processing import convert_to_negative as cn
from digital_image_processing import sepia as sp
from digital_image_processing.dithering import burkes as bs
from digital_image_processing.edge_detection import canny
from digital_image_processing.filters import convolve as conv
from digital_image_processing.filters import gaussian_filter as gg
from digital_image_processing.filters import local_binary_pattern as lbp
from digital_image_processing.filters import median_filter as med
from digital_image_processing.filters import sobel_filter as sob
from digital_image_processing.resize import resize as rs
__UpperCAmelCase = imread(R"digital_image_processing/image_data/lena_small.jpg")
__UpperCAmelCase = cvtColor(img, COLOR_BGR2GRAY)
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = cn.convert_to_negative(__A )
# assert negative_img array for at least one True
assert negative_img.any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
with Image.open('digital_image_processing/image_data/lena_small.jpg' ) as img:
# Work around assertion for response
assert str(cc.change_contrast(__A , 1_10 ) ).startswith(
'<PIL.Image.Image image mode=RGB size=100x100 at' )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Any = canny.gen_gaussian_kernel(9 , sigma=1.4 )
# Assert ambiguous array
assert resp.all()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Any = imread('digital_image_processing/image_data/lena_small.jpg' , 0 )
# assert ambiguous array for all == True
assert canny_img.all()
snake_case: Any = canny.canny(__A )
# assert canny array for at least one True
assert canny_array.any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert gg.gaussian_filter(__A , 5 , sigma=0.9 ).all()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Tuple = array([[0.25, 0.5, 0.25], [0.5, -3, 0.5], [0.25, 0.5, 0.25]] )
snake_case: int = conv.img_convolve(__A , __A ).astype(__A )
assert res.any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert med.median_filter(__A , 3 ).any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case , snake_case: str = sob.sobel_filter(__A )
assert grad.any() and theta.any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Any = sp.make_sepia(__A , 20 )
assert sepia.all()
def lowerCAmelCase_ ( __A : str = "digital_image_processing/image_data/lena_small.jpg" ):
'''simple docstring'''
snake_case: int = bs.Burkes(imread(__A , 1 ) , 1_20 )
burkes.process()
assert burkes.output_img.any()
def lowerCAmelCase_ ( __A : str = "digital_image_processing/image_data/lena_small.jpg" , ):
'''simple docstring'''
snake_case: List[Any] = rs.NearestNeighbour(imread(__A , 1 ) , 4_00 , 2_00 )
nn.process()
assert nn.output.any()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Optional[Any] = 'digital_image_processing/image_data/lena.jpg'
# Reading the image and converting it to grayscale.
snake_case: Optional[int] = imread(__A , 0 )
# Test for get_neighbors_pixel function() return not None
snake_case: Tuple = 0
snake_case: Optional[int] = 0
snake_case: Union[str, Any] = image[x_coordinate][y_coordinate]
snake_case: int = lbp.get_neighbors_pixel(
__A , __A , __A , __A )
assert neighbors_pixels is not None
# Test for local_binary_pattern function()
# Create a numpy array as the same height and width of read image
snake_case: Dict = np.zeros((image.shape[0], image.shape[1]) )
# Iterating through the image and calculating the local binary pattern value
# for each pixel.
for i in range(0 , image.shape[0] ):
for j in range(0 , image.shape[1] ):
snake_case: List[Any] = lbp.local_binary_value(__A , __A , __A )
assert lbp_image.any() | 692 |
'''simple docstring'''
import importlib
import os
import fsspec
import pytest
from fsspec import register_implementation
from fsspec.registry import _registry as _fsspec_registry
from datasets.filesystems import COMPRESSION_FILESYSTEMS, HfFileSystem, extract_path_from_uri, is_remote_filesystem
from .utils import require_lza, require_zstandard
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
assert "mock" in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
assert "mock" not in _fsspec_registry
assert "bz2" in _fsspec_registry
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'mock-s3-bucket'
snake_case: int = f"""s3://{mock_bucket}"""
snake_case: Any = extract_path_from_uri(__A )
assert dataset_path.startswith('s3://' ) is False
snake_case: Union[str, Any] = './local/path'
snake_case: Union[str, Any] = extract_path_from_uri(__A )
assert dataset_path == new_dataset_path
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
snake_case: List[str] = is_remote_filesystem(__A )
assert is_remote is True
snake_case: int = fsspec.filesystem('file' )
snake_case: int = is_remote_filesystem(__A )
assert is_remote is False
@pytest.mark.parametrize('compression_fs_class' , __A )
def lowerCAmelCase_ ( __A : Optional[int] , __A : int , __A : str , __A : Optional[Any] , __A : List[str] , __A : Optional[Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: Optional[Any] = {'gzip': gz_file, 'xz': xz_file, 'zstd': zstd_file, 'bz2': bza_file, 'lz4': lza_file}
snake_case: Optional[int] = input_paths[compression_fs_class.protocol]
if input_path is None:
snake_case: str = f"""for '{compression_fs_class.protocol}' compression protocol, """
if compression_fs_class.protocol == "lz4":
reason += require_lza.kwargs["reason"]
elif compression_fs_class.protocol == "zstd":
reason += require_zstandard.kwargs["reason"]
pytest.skip(__A )
snake_case: List[str] = fsspec.filesystem(compression_fs_class.protocol , fo=__A )
assert isinstance(__A , __A )
snake_case: Any = os.path.basename(__A )
snake_case: int = expected_filename[: expected_filename.rindex('.' )]
assert fs.glob('*' ) == [expected_filename]
with fs.open(__A , 'r' , encoding='utf-8' ) as f, open(__A , encoding='utf-8' ) as expected_file:
assert f.read() == expected_file.read()
@pytest.mark.parametrize('protocol' , ['zip', 'gzip'] )
def lowerCAmelCase_ ( __A : Any , __A : int , __A : int ):
'''simple docstring'''
snake_case: List[str] = {'zip': zip_jsonl_path, 'gzip': jsonl_gz_path}
snake_case: str = compressed_file_paths[protocol]
snake_case: Dict = 'dataset.jsonl'
snake_case: Optional[Any] = f"""{protocol}://{member_file_path}::{compressed_file_path}"""
snake_case , *snake_case: List[Any] = fsspec.get_fs_token_paths(__A )
assert fs.isfile(__A )
assert not fs.isfile('non_existing_' + member_file_path )
@pytest.mark.integration
def lowerCAmelCase_ ( __A : Optional[Any] , __A : Union[str, Any] , __A : Union[str, Any] , __A : List[Any] ):
'''simple docstring'''
snake_case: Tuple = hf_api.dataset_info(__A , token=__A )
snake_case: List[str] = HfFileSystem(repo_info=__A , token=__A )
assert sorted(hffs.glob('*' ) ) == [".gitattributes", "data"]
assert hffs.isdir('data' )
assert hffs.isfile('.gitattributes' ) and hffs.isfile('data/text_data.txt' )
with open(__A ) as f:
assert hffs.open('data/text_data.txt' , 'r' ).read() == f.read()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Union[str, Any] = 'bz2'
# Import module
import datasets.filesystems
# Overwrite protocol and reload
register_implementation(__A , __A , clobber=__A )
with pytest.warns(__A ) as warning_info:
importlib.reload(datasets.filesystems )
assert len(__A ) == 1
assert (
str(warning_info[0].message )
== f"""A filesystem protocol was already set for {protocol} and will be overwritten."""
) | 692 | 1 |
import argparse
import copy
def lowerCAmelCase_ ( __A : Dict ):
'''simple docstring'''
snake_case: Union[str, Any] = {}
with open(SCREAMING_SNAKE_CASE_ ) as f:
for line in f:
if line.split()[0] not in dict_of_neighbours:
snake_case: List[Any] = []
_list.append([line.split()[1], line.split()[2]] )
snake_case: int = _list
else:
dict_of_neighbours[line.split()[0]].append(
[line.split()[1], line.split()[2]] )
if line.split()[1] not in dict_of_neighbours:
snake_case: Dict = []
_list.append([line.split()[0], line.split()[2]] )
snake_case: Optional[Any] = _list
else:
dict_of_neighbours[line.split()[1]].append(
[line.split()[0], line.split()[2]] )
return dict_of_neighbours
def lowerCAmelCase_ ( __A : Union[str, Any] , __A : List[str] ):
'''simple docstring'''
with open(SCREAMING_SNAKE_CASE_ ) as f:
snake_case: List[Any] = f.read(1 )
snake_case: Union[str, Any] = start_node
snake_case: List[str] = []
snake_case: int = start_node
snake_case: List[Any] = 0
while visiting not in first_solution:
snake_case: List[str] = 1_00_00
for k in dict_of_neighbours[visiting]:
if int(k[1] ) < int(SCREAMING_SNAKE_CASE_ ) and k[0] not in first_solution:
snake_case: Dict = k[1]
snake_case: List[Any] = k[0]
first_solution.append(SCREAMING_SNAKE_CASE_ )
snake_case: Tuple = distance_of_first_solution + int(SCREAMING_SNAKE_CASE_ )
snake_case: List[str] = best_node
first_solution.append(SCREAMING_SNAKE_CASE_ )
snake_case: Optional[Any] = 0
for k in dict_of_neighbours[first_solution[-2]]:
if k[0] == start_node:
break
position += 1
snake_case: Optional[int] = (
distance_of_first_solution
+ int(dict_of_neighbours[first_solution[-2]][position][1] )
- 1_00_00
)
return first_solution, distance_of_first_solution
def lowerCAmelCase_ ( __A : Dict , __A : Dict ):
'''simple docstring'''
snake_case: Any = []
for n in solution[1:-1]:
snake_case: Dict = solution.index(SCREAMING_SNAKE_CASE_ )
for kn in solution[1:-1]:
snake_case: Optional[Any] = solution.index(SCREAMING_SNAKE_CASE_ )
if n == kn:
continue
snake_case: int = copy.deepcopy(SCREAMING_SNAKE_CASE_ )
snake_case: Dict = kn
snake_case: Optional[int] = n
snake_case: Optional[int] = 0
for k in _tmp[:-1]:
snake_case: str = _tmp[_tmp.index(SCREAMING_SNAKE_CASE_ ) + 1]
for i in dict_of_neighbours[k]:
if i[0] == next_node:
snake_case: int = distance + int(i[1] )
_tmp.append(SCREAMING_SNAKE_CASE_ )
if _tmp not in neighborhood_of_solution:
neighborhood_of_solution.append(_tmp )
snake_case: str = len(neighborhood_of_solution[0] ) - 1
neighborhood_of_solution.sort(key=lambda __A : x[index_of_last_item_in_the_list] )
return neighborhood_of_solution
def lowerCAmelCase_ ( __A : str , __A : Tuple , __A : Union[str, Any] , __A : Union[str, Any] , __A : Optional[int] ):
'''simple docstring'''
snake_case: int = 1
snake_case: Any = first_solution
snake_case: Optional[int] = []
snake_case: int = distance_of_first_solution
snake_case: Optional[Any] = solution
while count <= iters:
snake_case: Dict = find_neighborhood(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ )
snake_case: Tuple = 0
snake_case: List[Any] = neighborhood[index_of_best_solution]
snake_case: Tuple = len(SCREAMING_SNAKE_CASE_ ) - 1
snake_case: Any = False
while not found:
snake_case: List[str] = 0
while i < len(SCREAMING_SNAKE_CASE_ ):
if best_solution[i] != solution[i]:
snake_case: List[Any] = best_solution[i]
snake_case: Optional[Any] = solution[i]
break
snake_case: int = i + 1
if [first_exchange_node, second_exchange_node] not in tabu_list and [
second_exchange_node,
first_exchange_node,
] not in tabu_list:
tabu_list.append([first_exchange_node, second_exchange_node] )
snake_case: Optional[Any] = True
snake_case: Optional[Any] = best_solution[:-1]
snake_case: Any = neighborhood[index_of_best_solution][best_cost_index]
if cost < best_cost:
snake_case: int = cost
snake_case: Optional[int] = solution
else:
snake_case: Union[str, Any] = index_of_best_solution + 1
snake_case: Optional[int] = neighborhood[index_of_best_solution]
if len(SCREAMING_SNAKE_CASE_ ) >= size:
tabu_list.pop(0 )
snake_case: str = count + 1
return best_solution_ever, best_cost
def lowerCAmelCase_ ( __A : Optional[int]=None ):
'''simple docstring'''
snake_case: Any = generate_neighbours(args.File )
snake_case , snake_case: int = generate_first_solution(
args.File , SCREAMING_SNAKE_CASE_ )
snake_case , snake_case: Union[str, Any] = tabu_search(
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , args.Iterations , args.Size , )
print(f"""Best solution: {best_sol}, with total distance: {best_cost}.""" )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser(description="Tabu Search")
parser.add_argument(
"-f",
"--File",
type=str,
help="Path to the file containing the data",
required=True,
)
parser.add_argument(
"-i",
"--Iterations",
type=int,
help="How many iterations the algorithm should perform",
required=True,
)
parser.add_argument(
"-s", "--Size", type=int, help="Size of the tabu list", required=True
)
# Pass the arguments to main method
main(parser.parse_args()) | 700 |
'''simple docstring'''
import logging
import os
import random
import sys
from dataclasses import dataclass, field
from typing import Optional
import datasets
import numpy as np
import pandas as pd
from datasets import load_dataset
import transformers
from transformers import (
AutoConfig,
BartForSequenceClassification,
DataCollatorWithPadding,
EvalPrediction,
HfArgumentParser,
TapexTokenizer,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.17.0.dev0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
__UpperCAmelCase = logging.getLogger(__name__)
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The name of the dataset to use (via the datasets library)."} )
__UpperCamelCase = field(
default="tab_fact" , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} , )
__UpperCamelCase = field(
default=1024 , metadata={
"help": (
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Overwrite the cached preprocessed datasets or not."} )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Whether to pad all samples to `max_seq_length`. "
"If False, will pad the samples dynamically when batching to the maximum length in the batch."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"For debugging purposes or quicker training, truncate the number of prediction examples to this "
"value if set."
)
} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the training data."} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "A csv or a json file containing the validation data."} )
__UpperCamelCase = field(default=snake_case , metadata={"help": "A csv or a json file containing the test data."} )
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.dataset_name is not None:
pass
elif self.train_file is None or self.validation_file is None:
raise ValueError('Need either a GLUE task, a training/validation file or a dataset name.' )
else:
snake_case: str = self.train_file.split('.' )[-1]
assert train_extension in ["csv", "json"], "`train_file` should be a csv or a json file."
snake_case: Optional[Any] = self.validation_file.split('.' )[-1]
assert (
validation_extension == train_extension
), "`validation_file` should have the same extension (csv or json) as `train_file`."
@dataclass
class SCREAMING_SNAKE_CASE :
'''simple docstring'''
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"} , )
__UpperCamelCase = field(
default=snake_case , metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."} , )
__UpperCamelCase = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
__UpperCamelCase = field(
default=snake_case , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case , snake_case , snake_case: Union[str, Any] = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case , snake_case , snake_case: str = parser.parse_args_into_dataclasses()
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
snake_case: Tuple = training_args.get_process_log_level()
logger.setLevel(__A )
datasets.utils.logging.set_verbosity(__A )
transformers.utils.logging.set_verbosity(__A )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"""
+ f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" )
logger.info(f"""Training/evaluation parameters {training_args}""" )
# Detecting last checkpoint.
snake_case: Any = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case: List[Any] = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"""Output directory ({training_args.output_dir}) already exists and is not empty. """
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Set seed before initializing model.
set_seed(training_args.seed )
# Get the datasets: you can either provide your own CSV/JSON training and evaluation files (see below)
# or specify a GLUE benchmark task (the dataset will be downloaded automatically from the datasets Hub).
#
# For JSON files, this script will use the `question` column for the input question and `table` column for the corresponding table.
#
# If the CSVs/JSONs contain only one non-label column, the script does single sentence classification on this
# single column. You can easily tweak this behavior (see below)
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
snake_case: int = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from your local files.
# CSV/JSON training and evaluation files are needed.
snake_case: Optional[int] = {'train': data_args.train_file, 'validation': data_args.validation_file}
# Get the test dataset: you can provide your own CSV/JSON test file (see below)
# when you use `do_predict` without specifying a GLUE benchmark task.
if training_args.do_predict:
if data_args.test_file is not None:
snake_case: Tuple = data_args.train_file.split('.' )[-1]
snake_case: Union[str, Any] = data_args.test_file.split('.' )[-1]
assert (
test_extension == train_extension
), "`test_file` should have the same extension (csv or json) as `train_file`."
snake_case: Union[str, Any] = data_args.test_file
else:
raise ValueError('Need either a GLUE task or a test file for `do_predict`.' )
for key in data_files.keys():
logger.info(f"""load a local file for {key}: {data_files[key]}""" )
if data_args.train_file.endswith('.csv' ):
# Loading a dataset from local csv files
snake_case: List[Any] = load_dataset('csv' , data_files=__A , cache_dir=model_args.cache_dir )
else:
# Loading a dataset from local json files
snake_case: Optional[Any] = load_dataset('json' , data_files=__A , cache_dir=model_args.cache_dir )
# See more about loading any type of standard or custom dataset at
# https://huggingface.co/docs/datasets/loading_datasets.html.
# Labels
snake_case: Tuple = raw_datasets['train'].features['label'].names
snake_case: List[str] = len(__A )
# Load pretrained model and tokenizer
#
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case: Tuple = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# load tapex tokenizer
snake_case: List[str] = TapexTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , add_prefix_space=__A , )
snake_case: Union[str, Any] = BartForSequenceClassification.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=__A , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
# Padding strategy
if data_args.pad_to_max_length:
snake_case: int = 'max_length'
else:
# We will pad later, dynamically at batch creation, to the max sequence length in each batch
snake_case: Union[str, Any] = False
# Some models have set the order of the labels to use, so let's make sure we do use it.
snake_case: Optional[Any] = {'Refused': 0, 'Entailed': 1}
snake_case: List[Any] = {0: 'Refused', 1: 'Entailed'}
if data_args.max_seq_length > tokenizer.model_max_length:
logger.warning(
f"""The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"""
f"""model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}.""" )
snake_case: List[str] = min(data_args.max_seq_length , tokenizer.model_max_length )
def preprocess_tabfact_function(__A : Any ):
# Tokenize the texts
def _convert_table_text_to_pandas(__A : Dict ):
snake_case: str = [_table_row.split('#' ) for _table_row in _table_text.strip('\n' ).split('\n' )]
snake_case: List[str] = pd.DataFrame.from_records(_table_content[1:] , columns=_table_content[0] )
return _table_pd
snake_case: str = examples['statement']
snake_case: int = list(map(_convert_table_text_to_pandas , examples['table_text'] ) )
snake_case: List[Any] = tokenizer(__A , __A , padding=__A , max_length=__A , truncation=__A )
snake_case: List[Any] = examples['label']
return result
with training_args.main_process_first(desc='dataset map pre-processing' ):
snake_case: int = raw_datasets.map(
__A , batched=__A , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on dataset' , )
if training_args.do_train:
if "train" not in raw_datasets:
raise ValueError('--do_train requires a train dataset' )
snake_case: List[str] = raw_datasets['train']
if data_args.max_train_samples is not None:
snake_case: Tuple = train_dataset.select(range(data_args.max_train_samples ) )
if training_args.do_eval:
if "validation" not in raw_datasets and "validation_matched" not in raw_datasets:
raise ValueError('--do_eval requires a validation dataset' )
snake_case: Any = raw_datasets['validation']
if data_args.max_eval_samples is not None:
snake_case: Optional[int] = eval_dataset.select(range(data_args.max_eval_samples ) )
if training_args.do_predict or data_args.test_file is not None:
if "test" not in raw_datasets and "test_matched" not in raw_datasets:
raise ValueError('--do_predict requires a test dataset' )
snake_case: str = raw_datasets['test']
if data_args.max_predict_samples is not None:
snake_case: List[str] = predict_dataset.select(range(data_args.max_predict_samples ) )
# Log a few random samples from the training set:
if training_args.do_train:
for index in random.sample(range(len(__A ) ) , 3 ):
logger.info(f"""Sample {index} of the training set: {train_dataset[index]}.""" )
# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def compute_metrics(__A : EvalPrediction ):
snake_case: int = p.predictions[0] if isinstance(p.predictions , __A ) else p.predictions
snake_case: List[str] = np.argmax(__A , axis=1 )
return {"accuracy": (preds == p.label_ids).astype(np.floataa ).mean().item()}
# Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
if data_args.pad_to_max_length:
snake_case: str = default_data_collator
elif training_args.fpaa:
snake_case: List[str] = DataCollatorWithPadding(__A , pad_to_multiple_of=8 )
else:
snake_case: List[Any] = None
# Initialize our Trainer
snake_case: List[str] = Trainer(
model=__A , args=__A , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , compute_metrics=__A , tokenizer=__A , data_collator=__A , )
# Training
if training_args.do_train:
snake_case: Optional[int] = None
if training_args.resume_from_checkpoint is not None:
snake_case: str = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
snake_case: Optional[Any] = last_checkpoint
snake_case: Union[str, Any] = trainer.train(resume_from_checkpoint=__A )
snake_case: List[Any] = train_result.metrics
snake_case: List[Any] = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(__A )
)
snake_case: Optional[Any] = min(__A , len(__A ) )
trainer.save_model() # Saves the tokenizer too for easy upload
trainer.log_metrics('train' , __A )
trainer.save_metrics('train' , __A )
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info('*** Evaluate ***' )
snake_case: Dict = trainer.evaluate(eval_dataset=__A )
snake_case: Optional[int] = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(__A )
snake_case: Dict = min(__A , len(__A ) )
trainer.log_metrics('eval' , __A )
trainer.save_metrics('eval' , __A )
if training_args.do_predict:
logger.info('*** Predict ***' )
# Removing the `label` columns because it contains -1 and Trainer won't like that.
snake_case: Optional[int] = predict_dataset.remove_columns('label' )
snake_case: str = trainer.predict(__A , metric_key_prefix='predict' ).predictions
snake_case: Any = np.argmax(__A , axis=1 )
snake_case: int = os.path.join(training_args.output_dir , 'predict_results_tabfact.txt' )
if trainer.is_world_process_zero():
with open(__A , 'w' ) as writer:
logger.info('***** Predict Results *****' )
writer.write('index\tprediction\n' )
for index, item in enumerate(__A ):
snake_case: int = label_list[item]
writer.write(f"""{index}\t{item}\n""" )
snake_case: Optional[int] = {'finetuned_from': model_args.model_name_or_path, 'tasks': 'text-classification'}
if training_args.push_to_hub:
trainer.push_to_hub(**__A )
else:
trainer.create_model_card(**__A )
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
main()
if __name__ == "__main__":
main() | 692 | 0 |
'''simple docstring'''
import json
from typing import Iterator, List, Union
from tokenizers import AddedToken, Regex, Tokenizer, decoders, normalizers, pre_tokenizers, trainers
from tokenizers.implementations.base_tokenizer import BaseTokenizer
from tokenizers.models import Unigram
from tokenizers.processors import TemplateProcessing
class SCREAMING_SNAKE_CASE ( _snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = "▁" , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "<unk>" , SCREAMING_SNAKE_CASE__ = "</s>" , SCREAMING_SNAKE_CASE__ = "<pad>" , ):
'''simple docstring'''
snake_case: List[Any] = {
'pad': {'id': 0, 'token': pad_token},
'eos': {'id': 1, 'token': eos_token},
'unk': {'id': 2, 'token': unk_token},
}
snake_case: Optional[Any] = [None] * len(self.special_tokens )
for token_dict in self.special_tokens.values():
snake_case: int = token_dict['token']
snake_case: int = Tokenizer(Unigram() )
snake_case: Dict = normalizers.Sequence(
[
normalizers.Nmt(),
normalizers.NFKC(),
normalizers.Replace(Regex(' {2,}' ) , ' ' ),
normalizers.Lowercase(),
] )
snake_case: Any = pre_tokenizers.Sequence(
[
pre_tokenizers.Metaspace(replacement=lowerCAmelCase__ , add_prefix_space=lowerCAmelCase__ ),
pre_tokenizers.Digits(individual_digits=lowerCAmelCase__ ),
pre_tokenizers.Punctuation(),
] )
snake_case: List[Any] = decoders.Metaspace(replacement=lowerCAmelCase__ , add_prefix_space=lowerCAmelCase__ )
snake_case: int = TemplateProcessing(
single=F"""$A {self.special_tokens['eos']['token']}""" , special_tokens=[(self.special_tokens['eos']['token'], self.special_tokens['eos']['id'])] , )
snake_case: Tuple = {
'model': 'SentencePieceUnigram',
'replacement': replacement,
'add_prefix_space': add_prefix_space,
}
super().__init__(lowerCAmelCase__ , lowerCAmelCase__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = 80_00 , SCREAMING_SNAKE_CASE__ = True , ):
'''simple docstring'''
snake_case: int = trainers.UnigramTrainer(
vocab_size=lowerCAmelCase__ , special_tokens=self.special_tokens_list , show_progress=lowerCAmelCase__ , )
if isinstance(lowerCAmelCase__ , lowerCAmelCase__ ):
snake_case: Any = [files]
self._tokenizer.train(lowerCAmelCase__ , trainer=lowerCAmelCase__ )
self.add_unk_id()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = 80_00 , SCREAMING_SNAKE_CASE__ = True , ):
'''simple docstring'''
snake_case: Dict = trainers.UnigramTrainer(
vocab_size=lowerCAmelCase__ , special_tokens=self.special_tokens_list , show_progress=lowerCAmelCase__ , )
self._tokenizer.train_from_iterator(lowerCAmelCase__ , trainer=lowerCAmelCase__ )
self.add_unk_id()
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = json.loads(self._tokenizer.to_str() )
snake_case: Optional[int] = self.special_tokens['unk']['id']
snake_case: Union[str, Any] = Tokenizer.from_str(json.dumps(lowerCAmelCase__ ) ) | 701 |
'''simple docstring'''
import math
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or number % 2 == 0 or number % 3 == 0:
# Negatives, 0, 1, all even numbers, all multiples of 3 are not primes
return False
# All primes number are in format of 6k +/- 1
for i in range(5 , int(math.sqrt(__A ) + 1 ) , 6 ):
if number % i == 0 or number % (i + 2) == 0:
return False
return True
def lowerCAmelCase_ ( __A : float = 0.1 ):
'''simple docstring'''
snake_case: Optional[int] = 3
snake_case: int = 3
while primes / (2 * j - 1) >= ratio:
for i in range(j * j + j + 1 , (j + 2) * (j + 2) , j + 1 ):
primes += is_prime(__A )
j += 2
return j
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 0 |
'''simple docstring'''
import shutil
import tempfile
import unittest
import numpy as np
from transformers.testing_utils import (
is_pt_tf_cross_test,
require_tf,
require_torch,
require_torchvision,
require_vision,
)
from transformers.utils import is_tf_available, is_torch_available, is_vision_available
if is_vision_available():
from PIL import Image
from transformers import AutoProcessor, SamImageProcessor, SamProcessor
if is_torch_available():
import torch
if is_tf_available():
import tensorflow as tf
@require_vision
@require_torchvision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = tempfile.mkdtemp()
snake_case: Union[str, Any] = SamImageProcessor()
snake_case: int = SamProcessor(UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return AutoProcessor.from_pretrained(self.tmpdirname , **UpperCAmelCase_ ).image_processor
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Dict = [Image.fromarray(np.moveaxis(UpperCAmelCase_ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = SamProcessor(image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: List[Any] = self.get_image_processor(do_normalize=UpperCAmelCase_ , padding_value=1.0 )
snake_case: Optional[Any] = SamProcessor.from_pretrained(self.tmpdirname , do_normalize=UpperCAmelCase_ , padding_value=1.0 )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , UpperCAmelCase_ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_image_processor()
snake_case: Any = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: List[str] = self.prepare_image_inputs()
snake_case: int = image_processor(UpperCAmelCase_ , return_tensors='np' )
snake_case: Tuple = processor(images=UpperCAmelCase_ , return_tensors='np' )
input_feat_extract.pop('original_sizes' ) # pop original_sizes as it is popped in the processor
input_feat_extract.pop('reshaped_input_sizes' ) # pop original_sizes as it is popped in the processor
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_image_processor()
snake_case: str = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: List[Any] = [torch.ones((1, 3, 5, 5) )]
snake_case: Optional[Any] = [[17_64, 26_46]]
snake_case: Dict = [[6_83, 10_24]]
snake_case: Optional[Any] = processor.post_process_masks(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
snake_case: Tuple = processor.post_process_masks(
UpperCAmelCase_ , torch.tensor(UpperCAmelCase_ ) , torch.tensor(UpperCAmelCase_ ) )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
# should also work with np
snake_case: Any = [np.ones((1, 3, 5, 5) )]
snake_case: List[Any] = processor.post_process_masks(UpperCAmelCase_ , np.array(UpperCAmelCase_ ) , np.array(UpperCAmelCase_ ) )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
snake_case: str = [[1, 0], [0, 1]]
with self.assertRaises(UpperCAmelCase_ ):
snake_case: int = processor.post_process_masks(UpperCAmelCase_ , np.array(UpperCAmelCase_ ) , np.array(UpperCAmelCase_ ) )
@require_vision
@require_tf
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = SamImageProcessor()
snake_case: List[str] = SamProcessor(UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return AutoProcessor.from_pretrained(self.tmpdirname , **UpperCAmelCase_ ).image_processor
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Union[str, Any] = [Image.fromarray(np.moveaxis(UpperCAmelCase_ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = SamProcessor(image_processor=self.get_image_processor() )
processor.save_pretrained(self.tmpdirname )
snake_case: List[Any] = self.get_image_processor(do_normalize=UpperCAmelCase_ , padding_value=1.0 )
snake_case: str = SamProcessor.from_pretrained(self.tmpdirname , do_normalize=UpperCAmelCase_ , padding_value=1.0 )
self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() )
self.assertIsInstance(processor.image_processor , UpperCAmelCase_ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_image_processor()
snake_case: List[str] = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: Dict = self.prepare_image_inputs()
snake_case: List[str] = image_processor(UpperCAmelCase_ , return_tensors='np' )
snake_case: Any = processor(images=UpperCAmelCase_ , return_tensors='np' )
input_feat_extract.pop('original_sizes' ) # pop original_sizes as it is popped in the processor
input_feat_extract.pop('reshaped_input_sizes' ) # pop reshaped_input_sizes as it is popped in the processor
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_image_processor()
snake_case: Any = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: Optional[int] = [tf.ones((1, 3, 5, 5) )]
snake_case: int = [[17_64, 26_46]]
snake_case: Any = [[6_83, 10_24]]
snake_case: int = processor.post_process_masks(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , return_tensors='tf' )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
snake_case: List[str] = processor.post_process_masks(
UpperCAmelCase_ , tf.convert_to_tensor(UpperCAmelCase_ ) , tf.convert_to_tensor(UpperCAmelCase_ ) , return_tensors='tf' , )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
# should also work with np
snake_case: List[Any] = [np.ones((1, 3, 5, 5) )]
snake_case: Tuple = processor.post_process_masks(
UpperCAmelCase_ , np.array(UpperCAmelCase_ ) , np.array(UpperCAmelCase_ ) , return_tensors='tf' )
self.assertEqual(masks[0].shape , (1, 3, 17_64, 26_46) )
snake_case: Any = [[1, 0], [0, 1]]
with self.assertRaises(tf.errors.InvalidArgumentError ):
snake_case: Any = processor.post_process_masks(
UpperCAmelCase_ , np.array(UpperCAmelCase_ ) , np.array(UpperCAmelCase_ ) , return_tensors='tf' )
@require_vision
@require_torchvision
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = tempfile.mkdtemp()
snake_case: List[str] = SamImageProcessor()
snake_case: int = SamProcessor(UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return AutoProcessor.from_pretrained(self.tmpdirname , **UpperCAmelCase_ ).image_processor
def _UpperCamelCase ( self ):
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = [np.random.randint(2_55 , size=(3, 30, 4_00) , dtype=np.uinta )]
snake_case: Any = [Image.fromarray(np.moveaxis(UpperCAmelCase_ , 0 , -1 ) ) for x in image_inputs]
return image_inputs
@is_pt_tf_cross_test
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = self.get_image_processor()
snake_case: List[Any] = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: List[Any] = np.random.randint(0 , 2 , size=(1, 3, 5, 5) ).astype(np.floataa )
snake_case: int = [tf.convert_to_tensor(UpperCAmelCase_ )]
snake_case: List[Any] = [torch.tensor(UpperCAmelCase_ )]
snake_case: Dict = [[17_64, 26_46]]
snake_case: List[Any] = [[6_83, 10_24]]
snake_case: int = processor.post_process_masks(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , return_tensors='tf' )
snake_case: int = processor.post_process_masks(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , return_tensors='pt' )
self.assertTrue(np.all(tf_masks[0].numpy() == pt_masks[0].numpy() ) )
@is_pt_tf_cross_test
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = self.get_image_processor()
snake_case: Optional[Any] = SamProcessor(image_processor=UpperCAmelCase_ )
snake_case: Dict = self.prepare_image_inputs()
snake_case: Dict = image_processor(UpperCAmelCase_ , return_tensors='pt' )['pixel_values'].numpy()
snake_case: int = processor(images=UpperCAmelCase_ , return_tensors='pt' )['pixel_values'].numpy()
snake_case: Tuple = image_processor(UpperCAmelCase_ , return_tensors='tf' )['pixel_values'].numpy()
snake_case: Optional[Any] = processor(images=UpperCAmelCase_ , return_tensors='tf' )['pixel_values'].numpy()
self.assertTrue(np.allclose(UpperCAmelCase_ , UpperCAmelCase_ ) )
self.assertTrue(np.allclose(UpperCAmelCase_ , UpperCAmelCase_ ) )
self.assertTrue(np.allclose(UpperCAmelCase_ , UpperCAmelCase_ ) )
| 702 |
'''simple docstring'''
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCAmelCase = "pt"
elif is_tf_available():
__UpperCAmelCase = "tf"
else:
__UpperCAmelCase = "jax"
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = ByTaTokenizer
__UpperCamelCase = False
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: int = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname )
@cached_property
def _UpperCamelCase ( self ):
'''simple docstring'''
return ByTaTokenizer.from_pretrained('google/byt5-small' )
def _UpperCamelCase ( self , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.tokenizer_class.from_pretrained(self.tmpdirname , **SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=20 , SCREAMING_SNAKE_CASE__=5 ):
'''simple docstring'''
snake_case: Optional[Any] = []
for i in range(len(SCREAMING_SNAKE_CASE__ ) ):
try:
snake_case: Optional[Any] = tokenizer.decode([i] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
except UnicodeDecodeError:
pass
toks.append((i, tok) )
snake_case: List[str] = list(filter(lambda SCREAMING_SNAKE_CASE__ : re.match(r'^[ a-zA-Z]+$' , t[1] ) , SCREAMING_SNAKE_CASE__ ) )
snake_case: str = list(filter(lambda SCREAMING_SNAKE_CASE__ : [t[0]] == tokenizer.encode(t[1] , add_special_tokens=SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ ) )
if max_length is not None and len(SCREAMING_SNAKE_CASE__ ) > max_length:
snake_case: Union[str, Any] = toks[:max_length]
if min_length is not None and len(SCREAMING_SNAKE_CASE__ ) < min_length and len(SCREAMING_SNAKE_CASE__ ) > 0:
while len(SCREAMING_SNAKE_CASE__ ) < min_length:
snake_case: Tuple = toks + toks
# toks_str = [t[1] for t in toks]
snake_case: Dict = [t[0] for t in toks]
# Ensure consistency
snake_case: int = tokenizer.decode(SCREAMING_SNAKE_CASE__ , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
if " " not in output_txt and len(SCREAMING_SNAKE_CASE__ ) > 1:
snake_case: str = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=SCREAMING_SNAKE_CASE__ )
)
if with_prefix_space:
snake_case: Tuple = ' ' + output_txt
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
return output_txt, output_ids
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: str = tokenizer(['hi</s>', 'I went to the gym</s>', '</s>'] )
snake_case: List[Any] = tokenizer(['hi', 'I went to the gym', ''] )
self.assertListEqual(batch_with_eos_added['input_ids'] , batch_without_eos_added['input_ids'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.ta_base_tokenizer
snake_case: Union[str, Any] = 'Unicode €.'
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = [88, 1_13, 1_08, 1_02, 1_14, 1_03, 1_04, 35, 2_29, 1_33, 1_75, 49, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[str] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'Unicode €.</s>' )
snake_case: List[Any] = tokenizer('e è é ê ë' )
snake_case: Optional[Any] = [1_04, 35, 1_98, 1_71, 35, 1_98, 1_72, 35, 1_98, 1_73, 35, 1_98, 1_74, 1]
self.assertEqual(encoded['input_ids'] , SCREAMING_SNAKE_CASE__ )
# decoding
snake_case: List[Any] = tokenizer.decode(SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , 'e è é ê ë</s>' )
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë' ) ) , 'e è é ê ë</s>' )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = self.ta_base_tokenizer
snake_case: Optional[Any] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case: Optional[int] = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 1, 0]
# fmt: on
snake_case: str = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if FRAMEWORK != "jax":
snake_case: Optional[Any] = list(batch.input_ids.numpy()[0] )
else:
snake_case: Dict = list(batch.input_ids.tolist()[0] )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertEqual((2, 37) , batch.input_ids.shape )
self.assertEqual((2, 37) , batch.attention_mask.shape )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.ta_base_tokenizer
snake_case: List[str] = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case: Union[str, Any] = tokenizer(SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertIn('attention_mask' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_input_ids' , SCREAMING_SNAKE_CASE__ )
self.assertNotIn('decoder_attention_mask' , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.ta_base_tokenizer
snake_case: str = [
'Summary of the text.',
'Another summary.',
]
snake_case: Dict = tokenizer(
text_target=SCREAMING_SNAKE_CASE__ , max_length=32 , padding='max_length' , truncation=SCREAMING_SNAKE_CASE__ , return_tensors=SCREAMING_SNAKE_CASE__ )
self.assertEqual(32 , targets['input_ids'].shape[1] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = self.ta_base_tokenizer
snake_case: Optional[int] = ['A long paragraph for summarization. </s>']
snake_case: str = ['Summary of the text. </s>']
# fmt: off
snake_case: str = [68, 35, 1_11, 1_14, 1_13, 1_06, 35, 1_15, 1_00, 1_17, 1_00, 1_06, 1_17, 1_00, 1_15, 1_07, 35, 1_05, 1_14, 1_17, 35, 1_18, 1_20, 1_12, 1_12, 1_00, 1_17, 1_08, 1_25, 1_00, 1_19, 1_08, 1_14, 1_13, 49, 35, 1]
snake_case: Optional[int] = [86, 1_20, 1_12, 1_12, 1_00, 1_17, 1_24, 35, 1_14, 1_05, 35, 1_19, 1_07, 1_04, 35, 1_19, 1_04, 1_23, 1_19, 49, 35, 1]
# fmt: on
snake_case: List[Any] = tokenizer(SCREAMING_SNAKE_CASE__ , text_target=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['input_ids'][0] )
self.assertEqual(SCREAMING_SNAKE_CASE__ , batch['labels'][0] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
self.assertNotEqual(tokenizer.model_max_length , 42 )
# Now let's start the test
snake_case: Optional[Any] = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: Union[str, Any] = tempfile.mkdtemp()
snake_case: Dict = ' He is very happy, UNwant\u00E9d,running'
snake_case: Optional[int] = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Any = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.get_tokenizers(model_max_length=42 )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
# Isolate this from the other tests because we save additional tokens/etc
snake_case: List[str] = tempfile.mkdtemp()
snake_case: str = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'] )
snake_case: List[str] = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token' )
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens} )
snake_case: int = tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
tokenizer.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = after_tokenizer.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertIn('new_additional_special_token' , after_tokenizer.additional_special_tokens )
self.assertEqual(after_tokenizer.model_max_length , 42 )
snake_case: Union[str, Any] = tokenizer.__class__.from_pretrained(SCREAMING_SNAKE_CASE__ , model_max_length=43 )
self.assertEqual(tokenizer.model_max_length , 43 )
shutil.rmtree(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , encoding='utf-8' ) as json_file:
snake_case: Any = json.load(SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , encoding='utf-8' ) as json_file:
snake_case: str = json.load(SCREAMING_SNAKE_CASE__ )
snake_case: int = [F"""<extra_id_{i}>""" for i in range(1_25 )]
snake_case: Optional[int] = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case: str = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'special_tokens_map.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
with open(os.path.join(SCREAMING_SNAKE_CASE__ , 'tokenizer_config.json' ) , 'w' , encoding='utf-8' ) as outfile:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case: Dict = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , )
self.assertIn(
'an_additional_special_token' , tokenizer_without_change_in_init.additional_special_tokens )
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['an_additional_special_token'] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'] ) ) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case: Union[str, Any] = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token' , lstrip=SCREAMING_SNAKE_CASE__ )]
snake_case: Union[str, Any] = tokenizer_class.from_pretrained(
SCREAMING_SNAKE_CASE__ , additional_special_tokens=SCREAMING_SNAKE_CASE__ , )
self.assertIn('a_new_additional_special_token' , tokenizer.additional_special_tokens )
self.assertEqual(
['a_new_additional_special_token'] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'] ) ) , )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()) )
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()) )
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ )
self.assertTrue(tokenizer.decode([2_55] ) == '' )
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
pass
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.get_tokenizers(fast=SCREAMING_SNAKE_CASE__ , do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Union[str, Any] = ['t', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 'x', 't', '</s>']
snake_case: List[str] = tokenizer.convert_tokens_to_string(SCREAMING_SNAKE_CASE__ )
self.assertIsInstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Optional[Any] = [
'bos_token',
'eos_token',
'unk_token',
'sep_token',
'pad_token',
'cls_token',
'mask_token',
]
snake_case: Dict = 0
snake_case: List[Any] = tokenizer.convert_ids_to_tokens(
SCREAMING_SNAKE_CASE__ , skip_special_tokens=SCREAMING_SNAKE_CASE__ )
for attr in attributes_list:
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , attr + '_id' , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
self.assertEqual(getattr(SCREAMING_SNAKE_CASE__ , attr + '_id' ) , SCREAMING_SNAKE_CASE__ )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [] )
setattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' , [token_id_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens' ) , [token_to_test_setters] )
self.assertListEqual(getattr(SCREAMING_SNAKE_CASE__ , 'additional_special_tokens_ids' ) , [token_id_to_test_setters] ) | 692 | 0 |
'''simple docstring'''
from __future__ import annotations
from typing import TypedDict
class SCREAMING_SNAKE_CASE ( _SCREAMING_SNAKE_CASE ):
'''simple docstring'''
__UpperCamelCase = 42
__UpperCamelCase = 42
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
if not isinstance(__A , __A ):
raise TypeError('The parameter s type must be str.' )
return [s[i:] + s[:i] for i in range(len(__A ) )]
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
if not isinstance(__A , __A ):
raise TypeError('The parameter s type must be str.' )
if not s:
raise ValueError('The parameter s must not be empty.' )
snake_case: int = all_rotations(__A )
rotations.sort() # sort the list of rotations in alphabetically order
# make a string composed of the last char of each rotation
snake_case: Tuple = {
'bwt_string': ''.join([word[-1] for word in rotations] ),
'idx_original_string': rotations.index(__A ),
}
return response
def lowerCAmelCase_ ( __A : List[Any] , __A : Dict ):
'''simple docstring'''
if not isinstance(__A , __A ):
raise TypeError('The parameter bwt_string type must be str.' )
if not bwt_string:
raise ValueError('The parameter bwt_string must not be empty.' )
try:
snake_case: List[str] = int(__A )
except ValueError:
raise TypeError(
'The parameter idx_original_string type must be int or passive'
' of cast to int.' )
if idx_original_string < 0:
raise ValueError('The parameter idx_original_string must not be lower than 0.' )
if idx_original_string >= len(__A ):
raise ValueError(
'The parameter idx_original_string must be lower than' ' len(bwt_string).' )
snake_case: str = [''] * len(__A )
for _ in range(len(__A ) ):
for i in range(len(__A ) ):
snake_case: Optional[int] = bwt_string[i] + ordered_rotations[i]
ordered_rotations.sort()
return ordered_rotations[idx_original_string]
if __name__ == "__main__":
__UpperCAmelCase = "Provide a string that I will generate its BWT transform: "
__UpperCAmelCase = input(entry_msg).strip()
__UpperCAmelCase = bwt_transform(s)
print(
F'Burrows Wheeler transform for string \'{s}\' results '
F'in \'{result["bwt_string"]}\''
)
__UpperCAmelCase = reverse_bwt(result["bwt_string"], result["idx_original_string"])
print(
F'Reversing Burrows Wheeler transform for entry \'{result["bwt_string"]}\' '
F'we get original string \'{original_string}\''
) | 703 |
'''simple docstring'''
from typing import Any, Dict, Optional
import torch
import torch.nn.functional as F
from torch import nn
from ..utils import maybe_allow_in_graph
from .activations import get_activation
from .attention_processor import Attention
from .embeddings import CombinedTimestepLabelEmbeddings
@maybe_allow_in_graph
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=0.0 , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = False , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = "layer_norm" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = only_cross_attention
snake_case: Optional[Any] = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm_zero'
snake_case: Tuple = (num_embeds_ada_norm is not None) and norm_type == 'ada_norm'
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
raise ValueError(
F"""`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"""
F""" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}.""" )
# Define 3 blocks. Each block has its own normalization layer.
# 1. Self-Attn
if self.use_ada_layer_norm:
snake_case: List[str] = AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case: str = AdaLayerNormZero(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
else:
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if only_cross_attention else None , upcast_attention=SCREAMING_SNAKE_CASE__ , )
# 2. Cross-Attn
if cross_attention_dim is not None or double_self_attention:
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
# the second cross attention block.
snake_case: Tuple = (
AdaLayerNorm(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm
else nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = Attention(
query_dim=SCREAMING_SNAKE_CASE__ , cross_attention_dim=cross_attention_dim if not double_self_attention else None , heads=SCREAMING_SNAKE_CASE__ , dim_head=SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , bias=SCREAMING_SNAKE_CASE__ , upcast_attention=SCREAMING_SNAKE_CASE__ , ) # is self-attn if encoder_hidden_states is none
else:
snake_case: int = None
snake_case: Tuple = None
# 3. Feed-forward
snake_case: Union[str, Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = FeedForward(SCREAMING_SNAKE_CASE__ , dropout=SCREAMING_SNAKE_CASE__ , activation_fn=SCREAMING_SNAKE_CASE__ , final_dropout=SCREAMING_SNAKE_CASE__ )
# let chunk size default to None
snake_case: Any = None
snake_case: Any = 0
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = chunk_size
snake_case: str = dim
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
if self.use_ada_layer_norm:
snake_case: Optional[int] = self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.use_ada_layer_norm_zero:
snake_case , snake_case , snake_case , snake_case , snake_case: int = self.norma(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=hidden_states.dtype )
else:
snake_case: List[str] = self.norma(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = cross_attention_kwargs if cross_attention_kwargs is not None else {}
snake_case: List[str] = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=encoder_hidden_states if self.only_cross_attention else None , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
if self.use_ada_layer_norm_zero:
snake_case: Tuple = gate_msa.unsqueeze(1 ) * attn_output
snake_case: List[str] = attn_output + hidden_states
# 2. Cross-Attention
if self.attna is not None:
snake_case: Dict = (
self.norma(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) if self.use_ada_layer_norm else self.norma(SCREAMING_SNAKE_CASE__ )
)
snake_case: Any = self.attna(
SCREAMING_SNAKE_CASE__ , encoder_hidden_states=SCREAMING_SNAKE_CASE__ , attention_mask=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ , )
snake_case: List[str] = attn_output + hidden_states
# 3. Feed-forward
snake_case: str = self.norma(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: str = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
if self._chunk_size is not None:
# "feed_forward_chunk_size" can be used to save memory
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
raise ValueError(
F"""`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`.""" )
snake_case: List[str] = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
snake_case: Optional[Any] = torch.cat(
[self.ff(SCREAMING_SNAKE_CASE__ ) for hid_slice in norm_hidden_states.chunk(SCREAMING_SNAKE_CASE__ , dim=self._chunk_dim )] , dim=self._chunk_dim , )
else:
snake_case: int = self.ff(SCREAMING_SNAKE_CASE__ )
if self.use_ada_layer_norm_zero:
snake_case: Union[str, Any] = gate_mlp.unsqueeze(1 ) * ff_output
snake_case: Tuple = ff_output + hidden_states
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 4 , SCREAMING_SNAKE_CASE__ = 0.0 , SCREAMING_SNAKE_CASE__ = "geglu" , SCREAMING_SNAKE_CASE__ = False , ):
'''simple docstring'''
super().__init__()
snake_case: int = int(dim * mult )
snake_case: Optional[Any] = dim_out if dim_out is not None else dim
if activation_fn == "gelu":
snake_case: int = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if activation_fn == "gelu-approximate":
snake_case: Optional[Any] = GELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , approximate='tanh' )
elif activation_fn == "geglu":
snake_case: List[Any] = GEGLU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif activation_fn == "geglu-approximate":
snake_case: Optional[int] = ApproximateGELU(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.ModuleList([] )
# project in
self.net.append(SCREAMING_SNAKE_CASE__ )
# project dropout
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
# project out
self.net.append(nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) )
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
if final_dropout:
self.net.append(nn.Dropout(SCREAMING_SNAKE_CASE__ ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
for module in self.net:
snake_case: Optional[int] = module(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = "none" ):
'''simple docstring'''
super().__init__()
snake_case: Optional[int] = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = approximate
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ , approximate=self.approximate )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) , approximate=self.approximate ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.proj(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = self.gelu(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = nn.Linear(SCREAMING_SNAKE_CASE__ , dim_out * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if gate.device.type != "mps":
return F.gelu(SCREAMING_SNAKE_CASE__ )
# mps: gelu is not implemented for float16
return F.gelu(gate.to(dtype=torch.floataa ) ).to(dtype=gate.dtype )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case , snake_case: int = self.proj(SCREAMING_SNAKE_CASE__ ).chunk(2 , dim=-1 )
return hidden_states * self.gelu(SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.Linear(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = self.proj(SCREAMING_SNAKE_CASE__ )
return x * torch.sigmoid(1.7_02 * x )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Optional[Any] = nn.Embedding(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = nn.SiLU()
snake_case: Union[str, Any] = nn.Linear(SCREAMING_SNAKE_CASE__ , embedding_dim * 2 )
snake_case: int = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case: Dict = torch.chunk(SCREAMING_SNAKE_CASE__ , 2 )
snake_case: str = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale) + shift
return x
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = CombinedTimestepLabelEmbeddings(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.SiLU()
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , 6 * embedding_dim , bias=SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = nn.LayerNorm(SCREAMING_SNAKE_CASE__ , elementwise_affine=SCREAMING_SNAKE_CASE__ , eps=1E-6 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
snake_case: int = self.linear(self.silu(self.emb(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , hidden_dtype=SCREAMING_SNAKE_CASE__ ) ) )
snake_case , snake_case , snake_case , snake_case , snake_case , snake_case: str = emb.chunk(6 , dim=1 )
snake_case: Dict = self.norm(SCREAMING_SNAKE_CASE__ ) * (1 + scale_msa[:, None]) + shift_msa[:, None]
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = 1E-5 ):
'''simple docstring'''
super().__init__()
snake_case: str = num_groups
snake_case: str = eps
if act_fn is None:
snake_case: Dict = None
else:
snake_case: List[str] = get_activation(SCREAMING_SNAKE_CASE__ )
snake_case: Any = nn.Linear(SCREAMING_SNAKE_CASE__ , out_dim * 2 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.act:
snake_case: Optional[Any] = self.act(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.linear(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = emb[:, :, None, None]
snake_case , snake_case: List[Any] = emb.chunk(2 , dim=1 )
snake_case: Any = F.group_norm(SCREAMING_SNAKE_CASE__ , self.num_groups , eps=self.eps )
snake_case: Optional[int] = x * (1 + scale) + shift
return x | 692 | 0 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
if edge <= 0 or not isinstance(_lowercase , _lowercase ):
raise ValueError('Length must be a positive.' )
return 3 * ((25 + 10 * (5 ** (1 / 2))) ** (1 / 2)) * (edge**2)
def lowerCAmelCase_ ( __A : Any ):
'''simple docstring'''
if edge <= 0 or not isinstance(_lowercase , _lowercase ):
raise ValueError('Length must be a positive.' )
return ((15 + (7 * (5 ** (1 / 2)))) / 4) * (edge**3)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 704 |
'''simple docstring'''
import json
import os
import unittest
from transformers.models.roc_bert.tokenization_roc_bert import (
VOCAB_FILES_NAMES,
RoCBertBasicTokenizer,
RoCBertTokenizer,
RoCBertWordpieceTokenizer,
_is_control,
_is_punctuation,
_is_whitespace,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin, filter_non_english
@require_tokenizers
class SCREAMING_SNAKE_CASE ( snake_case , unittest.TestCase ):
'''simple docstring'''
__UpperCamelCase = RoCBertTokenizer
__UpperCamelCase = None
__UpperCamelCase = False
__UpperCamelCase = True
__UpperCamelCase = filter_non_english
def _UpperCamelCase ( self ):
'''simple docstring'''
super().setUp()
snake_case: Any = ['[UNK]', '[CLS]', '[SEP]', '[PAD]', '[MASK]', '你', '好', '是', '谁', 'a', 'b', 'c', 'd']
snake_case: List[Any] = {}
snake_case: List[str] = {}
for i, value in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = i
snake_case: Union[str, Any] = i
snake_case: List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Union[str, Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_shape_file'] )
snake_case: str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['word_pronunciation_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as vocab_writer:
vocab_writer.write(''.join([x + '\n' for x in vocab_tokens] ) )
with open(self.word_shape_file , 'w' , encoding='utf-8' ) as word_shape_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
with open(self.word_pronunciation_file , 'w' , encoding='utf-8' ) as word_pronunciation_writer:
json.dump(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , ensure_ascii=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: Dict = tokenizer.tokenize('你好[SEP]你是谁' )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , ['你', '好', '[SEP]', '你', '是', '谁'] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
self.assertListEqual(tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ ) , [5, 6, 2, 5, 7, 8] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer()
self.assertListEqual(tokenizer.tokenize('ah\u535A\u63A8zz' ) , ['ah', '\u535A', '\u63A8', 'zz'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['hello', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hällo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['h\u00E9llo'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['hallo', '!', 'how', 'are', 'you', '?'] )
self.assertListEqual(tokenizer.tokenize('H\u00E9llo' ) , ['hello'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? ' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HäLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , strip_accents=SCREAMING_SNAKE_CASE__ )
self.assertListEqual(
tokenizer.tokenize(' \tHäLLo!how \n Are yoU? ' ) , ['HaLLo', '!', 'how', 'Are', 'yoU', '?'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = RoCBertBasicTokenizer(do_lower_case=SCREAMING_SNAKE_CASE__ , never_split=['[UNK]'] )
self.assertListEqual(
tokenizer.tokenize(' \tHeLLo!how \n Are yoU? [UNK]' ) , ['HeLLo', '!', 'how', 'Are', 'yoU', '?', '[UNK]'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Union[str, Any] = ['[UNK]', '[CLS]', '[SEP]', 'want', '##want', '##ed', 'wa', 'un', 'runn', '##ing']
snake_case: Union[str, Any] = {}
for i, token in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: str = i
snake_case: Optional[int] = RoCBertWordpieceTokenizer(vocab=SCREAMING_SNAKE_CASE__ , unk_token='[UNK]' )
self.assertListEqual(tokenizer.tokenize('' ) , [] )
self.assertListEqual(tokenizer.tokenize('unwanted running' ) , ['un', '##want', '##ed', 'runn', '##ing'] )
self.assertListEqual(tokenizer.tokenize('unwantedX running' ) , ['[UNK]', 'runn', '##ing'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_whitespace(' ' ) )
self.assertTrue(_is_whitespace('\t' ) )
self.assertTrue(_is_whitespace('\r' ) )
self.assertTrue(_is_whitespace('\n' ) )
self.assertTrue(_is_whitespace('\u00A0' ) )
self.assertFalse(_is_whitespace('A' ) )
self.assertFalse(_is_whitespace('-' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_control('\u0005' ) )
self.assertFalse(_is_control('A' ) )
self.assertFalse(_is_control(' ' ) )
self.assertFalse(_is_control('\t' ) )
self.assertFalse(_is_control('\r' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
self.assertTrue(_is_punctuation('-' ) )
self.assertTrue(_is_punctuation('$' ) )
self.assertTrue(_is_punctuation('`' ) )
self.assertTrue(_is_punctuation('.' ) )
self.assertFalse(_is_punctuation('A' ) )
self.assertFalse(_is_punctuation(' ' ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = self.get_tokenizer()
# Example taken from the issue https://github.com/huggingface/tokenizers/issues/340
self.assertListEqual([tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
if self.test_rust_tokenizer:
snake_case: int = self.get_rust_tokenizer()
self.assertListEqual(
[rust_tokenizer.tokenize(SCREAMING_SNAKE_CASE__ ) for t in ['Test', '\xad', 'test']] , [['[UNK]'], [], ['[UNK]']] )
def _UpperCamelCase ( self ):
'''simple docstring'''
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = F"""A, naïve {tokenizer_r.mask_token} AllenNLP sentence."""
snake_case: List[str] = tokenizer_r.encode_plus(
SCREAMING_SNAKE_CASE__ , return_attention_mask=SCREAMING_SNAKE_CASE__ , return_token_type_ids=SCREAMING_SNAKE_CASE__ , return_offsets_mapping=SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ , )
snake_case: Optional[int] = tokenizer_r.do_lower_case if hasattr(SCREAMING_SNAKE_CASE__ , 'do_lower_case' ) else False
snake_case: int = (
[
((0, 0), tokenizer_r.cls_token),
((0, 1), 'A'),
((1, 2), ','),
((3, 5), 'na'),
((5, 6), '##ï'),
((6, 8), '##ve'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'Allen'),
((21, 23), '##NL'),
((23, 24), '##P'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
if not do_lower_case
else [
((0, 0), tokenizer_r.cls_token),
((0, 1), 'a'),
((1, 2), ','),
((3, 8), 'naive'),
((9, 15), tokenizer_r.mask_token),
((16, 21), 'allen'),
((21, 23), '##nl'),
((23, 24), '##p'),
((25, 33), 'sentence'),
((33, 34), '.'),
((0, 0), tokenizer_r.sep_token),
]
)
self.assertEqual(
[e[1] for e in expected_results] , tokenizer_r.convert_ids_to_tokens(tokens['input_ids'] ) )
self.assertEqual([e[0] for e in expected_results] , tokens['offset_mapping'] )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = ['的', '人', '有']
snake_case: Any = ''.join(SCREAMING_SNAKE_CASE__ )
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ):
snake_case: Tuple = True
snake_case: List[Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that each Chinese character is not preceded by "##"
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = False
snake_case: Union[str, Any] = self.rust_tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = self.tokenizer_class.from_pretrained(SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_r.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: int = tokenizer_p.encode(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer_r.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer_p.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ )
# it is expected that only the first Chinese character is not preceded by "##".
snake_case: Union[str, Any] = [
F"""##{token}""" if idx != 0 else token for idx, token in enumerate(SCREAMING_SNAKE_CASE__ )
]
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
self.assertListEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
@slow
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = self.tokenizer_class(self.vocab_file , self.word_shape_file , self.word_pronunciation_file )
snake_case: int = tokenizer.encode('你好' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Any = tokenizer.encode('你是谁' , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ )
snake_case: str = tokenizer.build_inputs_with_special_tokens(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
assert encoded_sentence == [1] + text + [2]
assert encoded_pair == [1] + text + [2] + text_a + [2]
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[str] = self.get_tokenizers(do_lower_case=SCREAMING_SNAKE_CASE__ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case: Dict = '你好，你是谁'
snake_case: int = tokenizer.tokenize(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_ids(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = tokenizer.convert_tokens_to_shape_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = tokenizer.convert_tokens_to_pronunciation_ids(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[Any] = tokenizer.prepare_for_model(
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = tokenizer.encode_plus(SCREAMING_SNAKE_CASE__ , add_special_tokens=SCREAMING_SNAKE_CASE__ )
self.assertEqual(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) | 692 | 0 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_sentencepiece_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_speecht5": [
"SPEECHT5_PRETRAINED_CONFIG_ARCHIVE_MAP",
"SPEECHT5_PRETRAINED_HIFIGAN_CONFIG_ARCHIVE_MAP",
"SpeechT5Config",
"SpeechT5HifiGanConfig",
],
"feature_extraction_speecht5": ["SpeechT5FeatureExtractor"],
"processing_speecht5": ["SpeechT5Processor"],
}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["SpeechT5Tokenizer"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"SPEECHT5_PRETRAINED_MODEL_ARCHIVE_LIST",
"SpeechT5ForSpeechToText",
"SpeechT5ForSpeechToSpeech",
"SpeechT5ForTextToSpeech",
"SpeechT5Model",
"SpeechT5PreTrainedModel",
"SpeechT5HifiGan",
]
if TYPE_CHECKING:
from .configuration_speechta import (
SPEECHT5_PRETRAINED_CONFIG_ARCHIVE_MAP,
SPEECHT5_PRETRAINED_HIFIGAN_CONFIG_ARCHIVE_MAP,
SpeechTaConfig,
SpeechTaHifiGanConfig,
)
from .feature_extraction_speechta import SpeechTaFeatureExtractor
from .processing_speechta import SpeechTaProcessor
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_speechta import SpeechTaTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_speechta import (
SPEECHT5_PRETRAINED_MODEL_ARCHIVE_LIST,
SpeechTaForSpeechToSpeech,
SpeechTaForSpeechToText,
SpeechTaForTextToSpeech,
SpeechTaHifiGan,
SpeechTaModel,
SpeechTaPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 705 |
'''simple docstring'''
from math import asin, atan, cos, radians, sin, sqrt, tan
__UpperCAmelCase = 6378137.0
__UpperCAmelCase = 6356752.314245
__UpperCAmelCase = 6_378_137
def lowerCAmelCase_ ( __A : float , __A : float , __A : float , __A : float ):
'''simple docstring'''
snake_case: Optional[Any] = (AXIS_A - AXIS_B) / AXIS_A
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: List[Any] = atan((1 - flattening) * tan(radians(__A ) ) )
snake_case: Tuple = radians(__A )
snake_case: Tuple = radians(__A )
# Equation
snake_case: List[Any] = sin((phi_a - phi_a) / 2 )
snake_case: Dict = sin((lambda_a - lambda_a) / 2 )
# Square both values
sin_sq_phi *= sin_sq_phi
sin_sq_lambda *= sin_sq_lambda
snake_case: Union[str, Any] = sqrt(sin_sq_phi + (cos(__A ) * cos(__A ) * sin_sq_lambda) )
return 2 * RADIUS * asin(__A )
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 0 |
'''simple docstring'''
from collections import OrderedDict
from typing import Mapping
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {
"distilbert-base-uncased": "https://huggingface.co/distilbert-base-uncased/resolve/main/config.json",
"distilbert-base-uncased-distilled-squad": (
"https://huggingface.co/distilbert-base-uncased-distilled-squad/resolve/main/config.json"
),
"distilbert-base-cased": "https://huggingface.co/distilbert-base-cased/resolve/main/config.json",
"distilbert-base-cased-distilled-squad": (
"https://huggingface.co/distilbert-base-cased-distilled-squad/resolve/main/config.json"
),
"distilbert-base-german-cased": "https://huggingface.co/distilbert-base-german-cased/resolve/main/config.json",
"distilbert-base-multilingual-cased": (
"https://huggingface.co/distilbert-base-multilingual-cased/resolve/main/config.json"
),
"distilbert-base-uncased-finetuned-sst-2-english": (
"https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english/resolve/main/config.json"
),
}
class SCREAMING_SNAKE_CASE ( __lowerCamelCase ):
'''simple docstring'''
__UpperCamelCase = "distilbert"
__UpperCamelCase = {
"hidden_size": "dim",
"num_attention_heads": "n_heads",
"num_hidden_layers": "n_layers",
}
def __init__( self , SCREAMING_SNAKE_CASE__=3_05_22 , SCREAMING_SNAKE_CASE__=5_12 , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=6 , SCREAMING_SNAKE_CASE__=12 , SCREAMING_SNAKE_CASE__=7_68 , SCREAMING_SNAKE_CASE__=4 * 7_68 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__="gelu" , SCREAMING_SNAKE_CASE__=0.02 , SCREAMING_SNAKE_CASE__=0.1 , SCREAMING_SNAKE_CASE__=0.2 , SCREAMING_SNAKE_CASE__=0 , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Dict = vocab_size
snake_case: Union[str, Any] = max_position_embeddings
snake_case: Dict = sinusoidal_pos_embds
snake_case: List[str] = n_layers
snake_case: Union[str, Any] = n_heads
snake_case: Any = dim
snake_case: Any = hidden_dim
snake_case: Union[str, Any] = dropout
snake_case: Union[str, Any] = attention_dropout
snake_case: Union[str, Any] = activation
snake_case: Optional[Any] = initializer_range
snake_case: str = qa_dropout
snake_case: str = seq_classif_dropout
super().__init__(**UpperCAmelCase_ , pad_token_id=UpperCAmelCase_ )
class SCREAMING_SNAKE_CASE ( __lowerCamelCase ):
'''simple docstring'''
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
if self.task == "multiple-choice":
snake_case: Union[str, Any] = {0: 'batch', 1: 'choice', 2: 'sequence'}
else:
snake_case: Union[str, Any] = {0: 'batch', 1: 'sequence'}
return OrderedDict(
[
('input_ids', dynamic_axis),
('attention_mask', dynamic_axis),
] ) | 706 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCAmelCase = {
"configuration_roformer": ["ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoFormerConfig", "RoFormerOnnxConfig"],
"tokenization_roformer": ["RoFormerTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = ["RoFormerTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoFormerForCausalLM",
"RoFormerForMaskedLM",
"RoFormerForMultipleChoice",
"RoFormerForQuestionAnswering",
"RoFormerForSequenceClassification",
"RoFormerForTokenClassification",
"RoFormerLayer",
"RoFormerModel",
"RoFormerPreTrainedModel",
"load_tf_weights_in_roformer",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFRoFormerForCausalLM",
"TFRoFormerForMaskedLM",
"TFRoFormerForMultipleChoice",
"TFRoFormerForQuestionAnswering",
"TFRoFormerForSequenceClassification",
"TFRoFormerForTokenClassification",
"TFRoFormerLayer",
"TFRoFormerModel",
"TFRoFormerPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCAmelCase = [
"FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
"FlaxRoFormerForMaskedLM",
"FlaxRoFormerForMultipleChoice",
"FlaxRoFormerForQuestionAnswering",
"FlaxRoFormerForSequenceClassification",
"FlaxRoFormerForTokenClassification",
"FlaxRoFormerModel",
"FlaxRoFormerPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig
from .tokenization_roformer import RoFormerTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_roformer_fast import RoFormerTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roformer import (
ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
RoFormerForCausalLM,
RoFormerForMaskedLM,
RoFormerForMultipleChoice,
RoFormerForQuestionAnswering,
RoFormerForSequenceClassification,
RoFormerForTokenClassification,
RoFormerLayer,
RoFormerModel,
RoFormerPreTrainedModel,
load_tf_weights_in_roformer,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_roformer import (
TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TFRoFormerForCausalLM,
TFRoFormerForMaskedLM,
TFRoFormerForMultipleChoice,
TFRoFormerForQuestionAnswering,
TFRoFormerForSequenceClassification,
TFRoFormerForTokenClassification,
TFRoFormerLayer,
TFRoFormerModel,
TFRoFormerPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_roformer import (
FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
FlaxRoFormerPreTrainedModel,
)
else:
import sys
__UpperCAmelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__) | 692 | 0 |
'''simple docstring'''
from math import factorial, pi
def lowerCAmelCase_ ( __A : float , __A : int = 30 ):
'''simple docstring'''
if not isinstance(__A , (int, float) ):
raise ValueError('maclaurin_sin() requires either an int or float for theta' )
if not isinstance(__A , __A ) or accuracy <= 0:
raise ValueError('maclaurin_sin() requires a positive int for accuracy' )
snake_case: Dict = float(__A )
snake_case: Dict = theta // (2 * pi)
theta -= 2 * div * pi
return sum(
(-1) ** r * theta ** (2 * r + 1) / factorial(2 * r + 1 ) for r in range(__A ) )
def lowerCAmelCase_ ( __A : float , __A : int = 30 ):
'''simple docstring'''
if not isinstance(__A , (int, float) ):
raise ValueError('maclaurin_cos() requires either an int or float for theta' )
if not isinstance(__A , __A ) or accuracy <= 0:
raise ValueError('maclaurin_cos() requires a positive int for accuracy' )
snake_case: int = float(__A )
snake_case: int = theta // (2 * pi)
theta -= 2 * div * pi
return sum((-1) ** r * theta ** (2 * r) / factorial(2 * r ) for r in range(__A ) )
if __name__ == "__main__":
import doctest
doctest.testmod()
print(maclaurin_sin(10))
print(maclaurin_sin(-10))
print(maclaurin_sin(10, 15))
print(maclaurin_sin(-10, 15))
print(maclaurin_cos(5))
print(maclaurin_cos(-5))
print(maclaurin_cos(10, 15))
print(maclaurin_cos(-10, 15)) | 707 |
'''simple docstring'''
import argparse
import torch
from datasets import load_dataset
from donut import DonutModel
from transformers import (
DonutImageProcessor,
DonutProcessor,
DonutSwinConfig,
DonutSwinModel,
MBartConfig,
MBartForCausalLM,
VisionEncoderDecoderModel,
XLMRobertaTokenizerFast,
)
def lowerCAmelCase_ ( __A : Tuple ):
'''simple docstring'''
snake_case: Tuple = model.config
snake_case: str = DonutSwinConfig(
image_size=original_config.input_size , patch_size=4 , depths=original_config.encoder_layer , num_heads=[4, 8, 16, 32] , window_size=original_config.window_size , embed_dim=1_28 , )
snake_case: Optional[Any] = MBartConfig(
is_decoder=__A , is_encoder_decoder=__A , add_cross_attention=__A , decoder_layers=original_config.decoder_layer , max_position_embeddings=original_config.max_position_embeddings , vocab_size=len(
model.decoder.tokenizer ) , scale_embedding=__A , add_final_layer_norm=__A , )
return encoder_config, decoder_config
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if "encoder.model" in name:
snake_case: Optional[Any] = name.replace('encoder.model' , 'encoder' )
if "decoder.model" in name:
snake_case: str = name.replace('decoder.model' , 'decoder' )
if "patch_embed.proj" in name:
snake_case: Any = name.replace('patch_embed.proj' , 'embeddings.patch_embeddings.projection' )
if "patch_embed.norm" in name:
snake_case: Optional[int] = name.replace('patch_embed.norm' , 'embeddings.norm' )
if name.startswith('encoder' ):
if "layers" in name:
snake_case: Tuple = 'encoder.' + name
if "attn.proj" in name:
snake_case: Optional[int] = name.replace('attn.proj' , 'attention.output.dense' )
if "attn" in name and "mask" not in name:
snake_case: Dict = name.replace('attn' , 'attention.self' )
if "norm1" in name:
snake_case: Union[str, Any] = name.replace('norm1' , 'layernorm_before' )
if "norm2" in name:
snake_case: Dict = name.replace('norm2' , 'layernorm_after' )
if "mlp.fc1" in name:
snake_case: List[str] = name.replace('mlp.fc1' , 'intermediate.dense' )
if "mlp.fc2" in name:
snake_case: Dict = name.replace('mlp.fc2' , 'output.dense' )
if name == "encoder.norm.weight":
snake_case: Dict = 'encoder.layernorm.weight'
if name == "encoder.norm.bias":
snake_case: int = 'encoder.layernorm.bias'
return name
def lowerCAmelCase_ ( __A : List[Any] , __A : Optional[Any] ):
'''simple docstring'''
for key in orig_state_dict.copy().keys():
snake_case: List[Any] = orig_state_dict.pop(__A )
if "qkv" in key:
snake_case: Union[str, Any] = key.split('.' )
snake_case: Optional[Any] = int(key_split[3] )
snake_case: Any = int(key_split[5] )
snake_case: Union[str, Any] = model.encoder.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
snake_case: Union[str, Any] = val[:dim, :]
snake_case: Any = val[dim : dim * 2, :]
snake_case: List[str] = val[-dim:, :]
else:
snake_case: str = val[:dim]
snake_case: Union[str, Any] = val[dim : dim * 2]
snake_case: List[Any] = val[-dim:]
elif "attn_mask" in key or key in ["encoder.model.norm.weight", "encoder.model.norm.bias"]:
# HuggingFace implementation doesn't use attn_mask buffer
# and model doesn't use final LayerNorms for the encoder
pass
else:
snake_case: Optional[int] = val
return orig_state_dict
def lowerCAmelCase_ ( __A : List[Any] , __A : Any=None , __A : List[str]=False ):
'''simple docstring'''
snake_case: str = DonutModel.from_pretrained(__A ).eval()
# load HuggingFace model
snake_case , snake_case: Optional[Any] = get_configs(__A )
snake_case: Optional[int] = DonutSwinModel(__A )
snake_case: Tuple = MBartForCausalLM(__A )
snake_case: Optional[Any] = VisionEncoderDecoderModel(encoder=__A , decoder=__A )
model.eval()
snake_case: Optional[int] = original_model.state_dict()
snake_case: Optional[int] = convert_state_dict(__A , __A )
model.load_state_dict(__A )
# verify results on scanned document
snake_case: Union[str, Any] = load_dataset('hf-internal-testing/example-documents' )
snake_case: str = dataset['test'][0]['image'].convert('RGB' )
snake_case: Optional[int] = XLMRobertaTokenizerFast.from_pretrained(__A , from_slow=__A )
snake_case: Any = DonutImageProcessor(
do_align_long_axis=original_model.config.align_long_axis , size=original_model.config.input_size[::-1] )
snake_case: Dict = DonutProcessor(__A , __A )
snake_case: Optional[Any] = processor(__A , return_tensors='pt' ).pixel_values
if model_name == "naver-clova-ix/donut-base-finetuned-docvqa":
snake_case: int = '<s_docvqa><s_question>{user_input}</s_question><s_answer>'
snake_case: Optional[Any] = 'When is the coffee break?'
snake_case: Optional[int] = task_prompt.replace('{user_input}' , __A )
elif model_name == "naver-clova-ix/donut-base-finetuned-rvlcdip":
snake_case: Dict = '<s_rvlcdip>'
elif model_name in [
"naver-clova-ix/donut-base-finetuned-cord-v1",
"naver-clova-ix/donut-base-finetuned-cord-v1-2560",
]:
snake_case: str = '<s_cord>'
elif model_name == "naver-clova-ix/donut-base-finetuned-cord-v2":
snake_case: str = 's_cord-v2>'
elif model_name == "naver-clova-ix/donut-base-finetuned-zhtrainticket":
snake_case: int = '<s_zhtrainticket>'
elif model_name in ["naver-clova-ix/donut-proto", "naver-clova-ix/donut-base"]:
# use a random prompt
snake_case: Optional[Any] = 'hello world'
else:
raise ValueError('Model name not supported' )
snake_case: Optional[int] = original_model.decoder.tokenizer(__A , add_special_tokens=__A , return_tensors='pt' )[
'input_ids'
]
snake_case: Any = original_model.encoder.model.patch_embed(__A )
snake_case , snake_case: Dict = model.encoder.embeddings(__A )
assert torch.allclose(__A , __A , atol=1E-3 )
# verify encoder hidden states
snake_case: Tuple = original_model.encoder(__A )
snake_case: List[str] = model.encoder(__A ).last_hidden_state
assert torch.allclose(__A , __A , atol=1E-2 )
# verify decoder hidden states
snake_case: List[Any] = original_model(__A , __A , __A ).logits
snake_case: List[Any] = model(__A , decoder_input_ids=__A ).logits
assert torch.allclose(__A , __A , atol=1E-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(f"""Saving model and processor to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
processor.save_pretrained(__A )
if push_to_hub:
model.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
processor.push_to_hub('nielsr/' + model_name.split('/' )[-1] , commit_message='Update model' )
if __name__ == "__main__":
__UpperCAmelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--model_name",
default="naver-clova-ix/donut-base-finetuned-docvqa",
required=False,
type=str,
help="Name of the original model you'd like to convert.",
)
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
required=False,
type=str,
help="Path to the output PyTorch model directory.",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether or not to push the converted model and processor to the 🤗 hub.",
)
__UpperCAmelCase = parser.parse_args()
convert_donut_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub) | 692 | 0 |
import argparse
import logging
import os
from datetime import datetime
import numpy as np
import torch
from torch import nn
from torch.utils.data import DataLoader, RandomSampler, TensorDataset
from tqdm import tqdm
from transformers import GPTaLMHeadModel
__UpperCAmelCase = logging.getLogger(__name__)
def lowerCAmelCase_ ( __A : str , __A : Any ):
'''simple docstring'''
if os.path.exists(__A ):
if os.path.exists(os.path.join(__A , 'config.json' ) ) and os.path.isfile(
os.path.join(__A , 'config.json' ) ):
os.remove(os.path.join(__A , 'config.json' ) )
if os.path.exists(os.path.join(__A , 'pytorch_model.bin' ) ) and os.path.isfile(
os.path.join(__A , 'pytorch_model.bin' ) ):
os.remove(os.path.join(__A , 'pytorch_model.bin' ) )
else:
os.makedirs(__A )
model.save_pretrained(__A )
def lowerCAmelCase_ ( __A : int , __A : Any=False ):
'''simple docstring'''
snake_case: Union[str, Any] = 2
if unlogit:
snake_case: str = torch.pow(__A , __A )
snake_case: str = p * torch.log(__A )
snake_case: List[str] = 0
return -plogp.sum(dim=-1 )
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
logger.info('lv, h >\t' + '\t'.join(f"""{x + 1}""" for x in range(len(__A ) ) ) )
for row in range(len(__A ) ):
if tensor.dtype != torch.long:
logger.info(f"""layer {row + 1}:\t""" + '\t'.join(f"""{x:.5f}""" for x in tensor[row].cpu().data ) )
else:
logger.info(f"""layer {row + 1}:\t""" + '\t'.join(f"""{x:d}""" for x in tensor[row].cpu().data ) )
def lowerCAmelCase_ ( __A : Tuple , __A : Union[str, Any] , __A : List[str] , __A : Optional[Any]=True , __A : int=True , __A : List[Any]=None , __A : int=False ):
'''simple docstring'''
snake_case: Dict = model.config.num_hidden_layers, model.config.num_attention_heads
snake_case: Optional[int] = torch.zeros(__A , __A ).to(args.device )
snake_case: Dict = torch.zeros(__A , __A ).to(args.device )
if head_mask is None:
snake_case: List[Any] = torch.ones(__A , __A ).to(args.device )
head_mask.requires_grad_(requires_grad=__A )
# If actually pruned attention multi-head, set head mask to None to avoid shape mismatch
if actually_pruned:
snake_case: int = None
snake_case: str = 0.0
snake_case: Dict = 0.0
for step, inputs in enumerate(tqdm(__A , desc='Iteration' , disable=args.local_rank not in [-1, 0] ) ):
snake_case: Dict = tuple(t.to(args.device ) for t in inputs )
(snake_case ): int = inputs
# Do a forward pass (not with torch.no_grad() since we need gradients for importance score - see below)
snake_case: Union[str, Any] = model(__A , labels=__A , head_mask=__A )
# (loss), lm_logits, presents, (all hidden_states), (attentions)
snake_case: int = (
outputs[0],
outputs[1],
outputs[-1],
) # Loss and logits are the first, attention the last
loss.backward() # Backpropagate to populate the gradients in the head mask
total_loss += loss.detach().cpu().numpy()
if compute_entropy:
for layer, attn in enumerate(__A ):
snake_case: Any = entropy(attn.detach() , __A )
attn_entropy[layer] += masked_entropy.sum(-1 ).sum(0 ).sum(0 ).detach()
if compute_importance:
head_importance += head_mask.grad.abs().detach()
tot_tokens += torch.ones_like(__A ).float().detach().sum().data
# Normalize
attn_entropy /= tot_tokens
head_importance /= tot_tokens
# Layerwise importance normalization
if not args.dont_normalize_importance_by_layer:
snake_case: Tuple = 2
snake_case: Any = torch.pow(torch.pow(__A , __A ).sum(-1 ) , 1 / exponent )
head_importance /= norm_by_layer.unsqueeze(-1 ) + 1E-20
if not args.dont_normalize_global_importance:
snake_case: List[str] = (head_importance - head_importance.min()) / (head_importance.max() - head_importance.min())
# Print matrices
if compute_entropy:
logger.info('Attention entropies' )
print_ad_tensor(__A )
if compute_importance:
logger.info('Head importance scores' )
print_ad_tensor(__A )
logger.info('Head ranked by importance scores' )
snake_case: Any = torch.zeros(head_importance.numel() , dtype=torch.long , device=args.device )
snake_case: Union[str, Any] = torch.arange(
head_importance.numel() , device=args.device )
snake_case: int = head_ranks.view_as(__A )
print_ad_tensor(__A )
return attn_entropy, head_importance, total_loss
def lowerCAmelCase_ ( __A : Any , __A : Dict , __A : Optional[Any] ):
'''simple docstring'''
snake_case: List[Any] = compute_heads_importance(__A , __A , __A , compute_entropy=__A )
snake_case: Tuple = 1 / loss # instead of downsteam score use the LM loss
logger.info('Pruning: original score: %f, threshold: %f' , __A , original_score * args.masking_threshold )
snake_case: Tuple = torch.ones_like(__A )
snake_case: str = max(1 , int(new_head_mask.numel() * args.masking_amount ) )
snake_case: Any = original_score
while current_score >= original_score * args.masking_threshold:
snake_case: Optional[int] = new_head_mask.clone().detach() # save current head mask
# heads from least important to most - keep only not-masked heads
snake_case: Dict = float('Inf' )
snake_case: List[Any] = head_importance.view(-1 ).sort()[1]
if len(__A ) <= num_to_mask:
print('BREAK BY num_to_mask' )
break
# mask heads
snake_case: List[Any] = current_heads_to_mask[:num_to_mask]
logger.info('Heads to mask: %s' , str(current_heads_to_mask.tolist() ) )
snake_case: Optional[int] = new_head_mask.view(-1 )
snake_case: Optional[Any] = 0.0
snake_case: Dict = new_head_mask.view_as(__A )
snake_case: Any = new_head_mask.clone().detach()
print_ad_tensor(__A )
# Compute metric and head importance again
snake_case: List[Any] = compute_heads_importance(
__A , __A , __A , compute_entropy=__A , head_mask=__A )
snake_case: Optional[int] = 1 / loss
logger.info(
'Masking: current score: %f, remaining heads %d (%.1f percents)' , __A , new_head_mask.sum() , new_head_mask.sum() / new_head_mask.numel() * 1_00 , )
logger.info('Final head mask' )
print_ad_tensor(__A )
np.save(os.path.join(args.output_dir , 'head_mask.npy' ) , head_mask.detach().cpu().numpy() )
return head_mask
def lowerCAmelCase_ ( __A : Dict , __A : int , __A : Dict , __A : int ):
'''simple docstring'''
snake_case: Optional[Any] = datetime.now()
snake_case: int = compute_heads_importance(
__A , __A , __A , compute_entropy=__A , compute_importance=__A , head_mask=__A )
snake_case: Tuple = 1 / loss
snake_case: Union[str, Any] = datetime.now() - before_time
snake_case: Dict = sum(p.numel() for p in model.parameters() )
snake_case: Tuple = {
layer: (1 - head_mask[layer].long()).nonzero().squeeze().tolist() for layer in range(len(__A ) )
}
for k, v in heads_to_prune.items():
if isinstance(__A , __A ):
snake_case: Optional[int] = [
v,
]
assert sum(len(__A ) for h in heads_to_prune.values() ) == (1 - head_mask.long()).sum().item()
model.prune_heads(__A )
snake_case: str = sum(p.numel() for p in model.parameters() )
snake_case: Optional[Any] = datetime.now()
snake_case: List[str] = compute_heads_importance(
__A , __A , __A , compute_entropy=__A , compute_importance=__A , head_mask=__A , actually_pruned=__A , )
snake_case: Any = 1 / loss
snake_case: List[str] = datetime.now() - before_time
logger.info(
'Pruning: original num of params: %.2e, after pruning %.2e (%.1f percents)' , __A , __A , pruned_num_params / original_num_params * 1_00 , )
logger.info('Pruning: score with masking: %f score with pruning: %f' , __A , __A )
logger.info('Pruning: speed ratio (original timing / new timing): %f percents' , original_time / new_time * 1_00 )
save_model(__A , args.output_dir )
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: str = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--data_dir' , default=__A , type=__A , required=__A , help='The input data dir. Should contain the .tsv files (or other data files) for the task.' , )
parser.add_argument(
'--model_name_or_path' , default=__A , type=__A , required=__A , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--output_dir' , default=__A , type=__A , required=__A , help='The output directory where the model predictions and checkpoints will be written.' , )
# Other parameters
parser.add_argument(
'--config_name' , default='' , type=__A , help='Pretrained config name or path if not the same as model_name_or_path' , )
parser.add_argument(
'--tokenizer_name' , default='' , type=__A , help='Pretrained tokenizer name or path if not the same as model_name_or_path' , )
parser.add_argument(
'--cache_dir' , default=__A , type=__A , help='Where do you want to store the pre-trained models downloaded from s3' , )
parser.add_argument(
'--data_subset' , type=__A , default=-1 , help='If > 0: limit the data to a subset of data_subset instances.' )
parser.add_argument(
'--overwrite_output_dir' , action='store_true' , help='Whether to overwrite data in output directory' )
parser.add_argument(
'--overwrite_cache' , action='store_true' , help='Overwrite the cached training and evaluation sets' )
parser.add_argument(
'--dont_normalize_importance_by_layer' , action='store_true' , help='Don\'t normalize importance score by layers' )
parser.add_argument(
'--dont_normalize_global_importance' , action='store_true' , help='Don\'t normalize all importance scores between 0 and 1' , )
parser.add_argument(
'--try_masking' , action='store_true' , help='Whether to try to mask head until a threshold of accuracy.' )
parser.add_argument(
'--masking_threshold' , default=0.9 , type=__A , help='masking threshold in term of metrics (stop masking when metric < threshold * original metric value).' , )
parser.add_argument(
'--masking_amount' , default=0.1 , type=__A , help='Amount to heads to masking at each masking step.' )
parser.add_argument('--metric_name' , default='acc' , type=__A , help='Metric to use for head masking.' )
parser.add_argument(
'--max_seq_length' , default=1_28 , type=__A , help=(
'The maximum total input sequence length after WordPiece tokenization. \n'
'Sequences longer than this will be truncated, sequences shorter padded.'
) , )
parser.add_argument('--batch_size' , default=1 , type=__A , help='Batch size.' )
parser.add_argument('--seed' , type=__A , default=42 )
parser.add_argument('--local_rank' , type=__A , default=-1 , help='local_rank for distributed training on gpus' )
parser.add_argument('--no_cuda' , action='store_true' , help='Whether not to use CUDA when available' )
parser.add_argument('--server_ip' , type=__A , default='' , help='Can be used for distant debugging.' )
parser.add_argument('--server_port' , type=__A , default='' , help='Can be used for distant debugging.' )
snake_case: str = parser.parse_args()
if args.server_ip and args.server_port:
# Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
import ptvsd
print('Waiting for debugger attach' )
ptvsd.enable_attach(address=(args.server_ip, args.server_port) , redirect_output=__A )
ptvsd.wait_for_attach()
# Setup devices and distributed training
if args.local_rank == -1 or args.no_cuda:
snake_case: Optional[Any] = torch.device('cuda' if torch.cuda.is_available() and not args.no_cuda else 'cpu' )
snake_case: Dict = 0 if args.no_cuda else torch.cuda.device_count()
else:
torch.cuda.set_device(args.local_rank )
snake_case: Dict = torch.device('cuda' , args.local_rank )
snake_case: Dict = 1
torch.distributed.init_process_group(backend='nccl' ) # Initializes the distributed backend
# Setup logging
logging.basicConfig(level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN )
logger.info('device: {} n_gpu: {}, distributed: {}'.format(args.device , args.n_gpu , bool(args.local_rank != -1 ) ) )
snake_case: int = GPTaLMHeadModel.from_pretrained(args.model_name_or_path )
# Distributed and parallel training
model.to(args.device )
if args.local_rank != -1:
snake_case: Union[str, Any] = nn.parallel.DistributedDataParallel(
__A , device_ids=[args.local_rank] , output_device=args.local_rank , find_unused_parameters=__A )
elif args.n_gpu > 1:
snake_case: Any = nn.DataParallel(__A )
# Print/save training arguments
os.makedirs(args.output_dir , exist_ok=__A )
torch.save(__A , os.path.join(args.output_dir , 'run_args.bin' ) )
logger.info('Training/evaluation parameters %s' , __A )
# Prepare dataset
snake_case: Union[str, Any] = np.concatenate(
[
np.loadtxt(args.data_dir , dtype=np.intaa ),
] )
snake_case: Dict = (torch.from_numpy(__A ),)
snake_case: Dict = TensorDataset(*__A )
snake_case: Optional[Any] = RandomSampler(__A )
snake_case: List[Any] = DataLoader(__A , sampler=__A , batch_size=args.batch_size )
# Compute head entropy and importance score
compute_heads_importance(__A , __A , __A )
# Try head masking (set heads to zero until the score goes under a threshole)
# and head pruning (remove masked heads and see the effect on the network)
if args.try_masking and args.masking_threshold > 0.0 and args.masking_threshold < 1.0:
snake_case: Tuple = mask_heads(__A , __A , __A )
prune_heads(__A , __A , __A , __A )
if __name__ == "__main__":
main() | 708 |
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class SCREAMING_SNAKE_CASE ( unittest.TestCase ):
'''simple docstring'''
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 1_28, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 1_42, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
snake_case: Union[str, Any] = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 1_28,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 1_42,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(SCREAMING_SNAKE_CASE__ ) , SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Dict = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , x.transpose() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Tuple = np.random.randn(3 , 4 )
snake_case: Optional[Any] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(3 , 4 , 5 )
snake_case: Optional[int] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , transpose(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Dict = np.random.randn(3 , 4 , 5 )
snake_case: str = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Optional[int] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Optional[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) , np.asarray(transpose(SCREAMING_SNAKE_CASE__ , axes=(1, 2, 0) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) )
snake_case: Optional[int] = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: List[str] = np.random.randn(3 , 4 , 5 )
snake_case: Tuple = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: List[Any] = np.random.randn(3 , 4 )
snake_case: Tuple = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ).numpy() ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: str = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (4, 3) ) ) ) )
snake_case: Any = np.random.randn(3 , 4 , 5 )
snake_case: List[str] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) , np.asarray(reshape(SCREAMING_SNAKE_CASE__ , (12, 5) ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.squeeze(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: List[str] = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(1 , 3 , 4 )
snake_case: List[str] = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Optional[Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: int = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Any = np.random.randn(1 , 3 , 4 )
snake_case: Optional[Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , squeeze(SCREAMING_SNAKE_CASE__ ).numpy() ) )
snake_case: Union[str, Any] = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Union[str, Any] = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(1 , 3 , 4 )
snake_case: List[Any] = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ ) ) ) )
snake_case: Tuple = np.random.randn(1 , 4 , 1 , 5 )
snake_case: Tuple = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) , np.asarray(squeeze(SCREAMING_SNAKE_CASE__ , axis=2 ) ) ) )
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) )
@require_torch
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: int = np.random.randn(3 , 4 )
snake_case: Any = torch.tensor(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_tf
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: str = np.random.randn(3 , 4 )
snake_case: Any = tf.constant(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ).numpy() ) )
@require_flax
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[Any] = np.random.randn(3 , 4 )
snake_case: int = jnp.array(SCREAMING_SNAKE_CASE__ )
self.assertTrue(np.allclose(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) , np.asarray(expand_dims(SCREAMING_SNAKE_CASE__ , axis=1 ) ) ) ) | 692 | 0 |
'''simple docstring'''
from typing import Dict, List, Optional, Union
import numpy as np
from transformers.utils import is_vision_available
from transformers.utils.generic import TensorType
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import logging
if is_vision_available():
import PIL
__UpperCAmelCase = logging.get_logger(__name__)
def lowerCAmelCase_ ( __A : int ):
'''simple docstring'''
if isinstance(__lowerCAmelCase , (list, tuple) ) and isinstance(videos[0] , (list, tuple) ) and is_valid_image(videos[0][0] ):
return videos
elif isinstance(__lowerCAmelCase , (list, tuple) ) and is_valid_image(videos[0] ):
return [videos]
elif is_valid_image(__lowerCAmelCase ):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""" )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = ["pixel_values"]
def __init__( self , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = PILImageResampling.BILINEAR , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = 1 / 2_55 , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
super().__init__(**_lowerCAmelCase )
snake_case: Optional[Any] = size if size is not None else {'shortest_edge': 2_56}
snake_case: int = get_size_dict(_lowerCAmelCase , default_to_square=_lowerCAmelCase )
snake_case: Tuple = crop_size if crop_size is not None else {'height': 2_24, 'width': 2_24}
snake_case: List[Any] = get_size_dict(_lowerCAmelCase , param_name='crop_size' )
snake_case: Dict = do_resize
snake_case: Tuple = size
snake_case: str = do_center_crop
snake_case: Optional[int] = crop_size
snake_case: Optional[int] = resample
snake_case: Optional[int] = do_rescale
snake_case: List[str] = rescale_factor
snake_case: str = offset
snake_case: int = do_normalize
snake_case: Optional[int] = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
snake_case: Optional[Any] = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = PILImageResampling.BILINEAR , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Union[str, Any] = get_size_dict(_lowerCAmelCase , default_to_square=_lowerCAmelCase )
if "shortest_edge" in size:
snake_case: List[Any] = get_resize_output_image_size(_lowerCAmelCase , size['shortest_edge'] , default_to_square=_lowerCAmelCase )
elif "height" in size and "width" in size:
snake_case: List[str] = (size['height'], size['width'])
else:
raise ValueError(F"""Size must have \'height\' and \'width\' or \'shortest_edge\' as keys. Got {size.keys()}""" )
return resize(_lowerCAmelCase , size=_lowerCAmelCase , resample=_lowerCAmelCase , data_format=_lowerCAmelCase , **_lowerCAmelCase )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: List[Any] = get_size_dict(_lowerCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(F"""Size must have \'height\' and \'width\' as keys. Got {size.keys()}""" )
return center_crop(_lowerCAmelCase , size=(size['height'], size['width']) , data_format=_lowerCAmelCase , **_lowerCAmelCase )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = True , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: List[Any] = image.astype(np.floataa )
if offset:
snake_case: str = image - (scale / 2)
return rescale(_lowerCAmelCase , scale=_lowerCAmelCase , data_format=_lowerCAmelCase , **_lowerCAmelCase )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
return normalize(_lowerCAmelCase , mean=_lowerCAmelCase , std=_lowerCAmelCase , data_format=_lowerCAmelCase , **_lowerCAmelCase )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = ChannelDimension.FIRST , ):
'''simple docstring'''
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
if offset and not do_rescale:
raise ValueError('For offset, do_rescale must also be set to True.' )
# All transformations expect numpy arrays.
snake_case: Union[str, Any] = to_numpy_array(_lowerCAmelCase )
if do_resize:
snake_case: str = self.resize(image=_lowerCAmelCase , size=_lowerCAmelCase , resample=_lowerCAmelCase )
if do_center_crop:
snake_case: int = self.center_crop(_lowerCAmelCase , size=_lowerCAmelCase )
if do_rescale:
snake_case: Union[str, Any] = self.rescale(image=_lowerCAmelCase , scale=_lowerCAmelCase , offset=_lowerCAmelCase )
if do_normalize:
snake_case: Union[str, Any] = self.normalize(image=_lowerCAmelCase , mean=_lowerCAmelCase , std=_lowerCAmelCase )
snake_case: str = to_channel_dimension_format(_lowerCAmelCase , _lowerCAmelCase )
return image
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = ChannelDimension.FIRST , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[int] = do_resize if do_resize is not None else self.do_resize
snake_case: Optional[int] = resample if resample is not None else self.resample
snake_case: Tuple = do_center_crop if do_center_crop is not None else self.do_center_crop
snake_case: str = do_rescale if do_rescale is not None else self.do_rescale
snake_case: Union[str, Any] = rescale_factor if rescale_factor is not None else self.rescale_factor
snake_case: int = offset if offset is not None else self.offset
snake_case: Tuple = do_normalize if do_normalize is not None else self.do_normalize
snake_case: Union[str, Any] = image_mean if image_mean is not None else self.image_mean
snake_case: List[Any] = image_std if image_std is not None else self.image_std
snake_case: int = size if size is not None else self.size
snake_case: Optional[int] = get_size_dict(_lowerCAmelCase , default_to_square=_lowerCAmelCase )
snake_case: str = crop_size if crop_size is not None else self.crop_size
snake_case: List[str] = get_size_dict(_lowerCAmelCase , param_name='crop_size' )
if not valid_images(_lowerCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
snake_case: str = make_batched(_lowerCAmelCase )
snake_case: int = [
[
self._preprocess_image(
image=_lowerCAmelCase , do_resize=_lowerCAmelCase , size=_lowerCAmelCase , resample=_lowerCAmelCase , do_center_crop=_lowerCAmelCase , crop_size=_lowerCAmelCase , do_rescale=_lowerCAmelCase , rescale_factor=_lowerCAmelCase , offset=_lowerCAmelCase , do_normalize=_lowerCAmelCase , image_mean=_lowerCAmelCase , image_std=_lowerCAmelCase , data_format=_lowerCAmelCase , )
for img in video
]
for video in videos
]
snake_case: Union[str, Any] = {'pixel_values': videos}
return BatchFeature(data=_lowerCAmelCase , tensor_type=_lowerCAmelCase ) | 709 |
'''simple docstring'''
import collections.abc
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
from ...activations import ACTaFN
from ...modeling_outputs import BaseModelOutputWithNoAttention, ImageClassifierOutputWithNoAttention
from ...modeling_utils import PreTrainedModel
from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
from .configuration_poolformer import PoolFormerConfig
__UpperCAmelCase = logging.get_logger(__name__)
# General docstring
__UpperCAmelCase = "PoolFormerConfig"
# Base docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = [1, 512, 7, 7]
# Image classification docstring
__UpperCAmelCase = "sail/poolformer_s12"
__UpperCAmelCase = "tabby, tabby cat"
__UpperCAmelCase = [
"sail/poolformer_s12",
# See all PoolFormer models at https://huggingface.co/models?filter=poolformer
]
def lowerCAmelCase_ ( __A : Tuple , __A : float = 0.0 , __A : bool = False ):
'''simple docstring'''
if drop_prob == 0.0 or not training:
return input
snake_case: Union[str, Any] = 1 - drop_prob
snake_case: List[Any] = (input.shape[0],) + (1,) * (input.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
snake_case: List[Any] = keep_prob + torch.rand(__A , dtype=input.dtype , device=input.device )
random_tensor.floor_() # binarize
snake_case: Any = input.div(__A ) * random_tensor
return output
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = drop_prob
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return drop_path(SCREAMING_SNAKE_CASE__ , self.drop_prob , self.training )
def _UpperCamelCase ( self ):
'''simple docstring'''
return "p={}".format(self.drop_prob )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=None ):
'''simple docstring'''
super().__init__()
snake_case: List[str] = patch_size if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (patch_size, patch_size)
snake_case: List[str] = stride if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (stride, stride)
snake_case: Union[str, Any] = padding if isinstance(SCREAMING_SNAKE_CASE__ , collections.abc.Iterable ) else (padding, padding)
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , kernel_size=SCREAMING_SNAKE_CASE__ , stride=SCREAMING_SNAKE_CASE__ , padding=SCREAMING_SNAKE_CASE__ )
snake_case: Tuple = norm_layer(SCREAMING_SNAKE_CASE__ ) if norm_layer else nn.Identity()
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.projection(SCREAMING_SNAKE_CASE__ )
snake_case: Any = self.norm(SCREAMING_SNAKE_CASE__ )
return embeddings
class SCREAMING_SNAKE_CASE ( nn.GroupNorm ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(1 , SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: str = nn.AvgPoolad(SCREAMING_SNAKE_CASE__ , stride=1 , padding=pool_size // 2 , count_include_pad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.pool(SCREAMING_SNAKE_CASE__ ) - hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: Any = nn.Convad(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , 1 )
snake_case: str = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ )
if isinstance(config.hidden_act , SCREAMING_SNAKE_CASE__ ):
snake_case: Tuple = ACTaFN[config.hidden_act]
else:
snake_case: int = config.hidden_act
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Dict = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = self.act_fn(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.drop(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = self.conva(SCREAMING_SNAKE_CASE__ )
snake_case: str = self.drop(SCREAMING_SNAKE_CASE__ )
return hidden_states
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Tuple = PoolFormerPooling(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerOutput(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
snake_case: Dict = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = PoolFormerGroupNorm(SCREAMING_SNAKE_CASE__ )
# Useful for training neural nets
snake_case: Union[str, Any] = PoolFormerDropPath(SCREAMING_SNAKE_CASE__ ) if drop_path > 0.0 else nn.Identity()
snake_case: Optional[Any] = config.use_layer_scale
if config.use_layer_scale:
snake_case: Any = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
snake_case: int = nn.Parameter(
config.layer_scale_init_value * torch.ones((SCREAMING_SNAKE_CASE__) ) , requires_grad=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if self.use_layer_scale:
snake_case: str = self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Dict = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * pooling_output
# First residual connection
snake_case: str = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: Dict = ()
snake_case: Dict = self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) )
snake_case: Union[str, Any] = self.layer_scale_a.unsqueeze(-1 ).unsqueeze(-1 ) * layer_output
# Second residual connection
snake_case: Any = hidden_states + self.drop_path(SCREAMING_SNAKE_CASE__ )
snake_case: List[str] = (output,) + outputs
return outputs
else:
snake_case: Optional[Any] = self.drop_path(self.pooling(self.before_norm(SCREAMING_SNAKE_CASE__ ) ) )
# First residual connection
snake_case: Union[str, Any] = pooling_output + hidden_states
snake_case: List[Any] = ()
# Second residual connection inside the PoolFormerOutput block
snake_case: List[str] = self.drop_path(self.output(self.after_norm(SCREAMING_SNAKE_CASE__ ) ) )
snake_case: Dict = hidden_states + layer_output
snake_case: Optional[Any] = (output,) + outputs
return outputs
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: List[Any] = config
# stochastic depth decay rule
snake_case: List[Any] = [x.item() for x in torch.linspace(0 , config.drop_path_rate , sum(config.depths ) )]
# patch embeddings
snake_case: Union[str, Any] = []
for i in range(config.num_encoder_blocks ):
embeddings.append(
PoolFormerEmbeddings(
patch_size=config.patch_sizes[i] , stride=config.strides[i] , padding=config.padding[i] , num_channels=config.num_channels if i == 0 else config.hidden_sizes[i - 1] , hidden_size=config.hidden_sizes[i] , ) )
snake_case: List[Any] = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
# Transformer blocks
snake_case: str = []
snake_case: int = 0
for i in range(config.num_encoder_blocks ):
# each block consists of layers
snake_case: List[str] = []
if i != 0:
cur += config.depths[i - 1]
for j in range(config.depths[i] ):
layers.append(
PoolFormerLayer(
SCREAMING_SNAKE_CASE__ , num_channels=config.hidden_sizes[i] , pool_size=config.pool_size , hidden_size=config.hidden_sizes[i] , intermediate_size=int(config.hidden_sizes[i] * config.mlp_ratio ) , drop_path=dpr[cur + j] , ) )
blocks.append(nn.ModuleList(SCREAMING_SNAKE_CASE__ ) )
snake_case: Tuple = nn.ModuleList(SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__=True ):
'''simple docstring'''
snake_case: str = () if output_hidden_states else None
snake_case: Dict = pixel_values
for idx, layers in enumerate(zip(self.patch_embeddings , self.block ) ):
snake_case , snake_case: Dict = layers
# Get patch embeddings from hidden_states
snake_case: int = embedding_layer(SCREAMING_SNAKE_CASE__ )
# Send the embeddings through the blocks
for _, blk in enumerate(SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = blk(SCREAMING_SNAKE_CASE__ )
snake_case: Optional[int] = layer_outputs[0]
if output_hidden_states:
snake_case: List[str] = all_hidden_states + (hidden_states,)
if not return_dict:
return tuple(v for v in [hidden_states, all_hidden_states] if v is not None )
return BaseModelOutputWithNoAttention(last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=SCREAMING_SNAKE_CASE__ )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = PoolFormerConfig
__UpperCamelCase = "poolformer"
__UpperCamelCase = "pixel_values"
__UpperCamelCase = True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , (nn.Linear, nn.Convad) ):
module.weight.data.normal_(mean=0.0 , std=self.config.initializer_range )
if module.bias is not None:
module.bias.data.zero_()
elif isinstance(SCREAMING_SNAKE_CASE__ , nn.LayerNorm ):
module.bias.data.zero_()
module.weight.data.fill_(1.0 )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False ):
'''simple docstring'''
if isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
snake_case: List[Any] = value
__UpperCAmelCase = R"\n This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) sub-class. Use\n it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and\n behavior.\n\n Parameters:\n config ([`PoolFormerConfig`]): Model configuration class with all the parameters of the model.\n Initializing with a config file does not load the weights associated with the model, only the\n configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.\n"
__UpperCAmelCase = R"\n Args:\n pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):\n Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See\n [`PoolFormerImageProcessor.__call__`] for details.\n"
@add_start_docstrings(
"The bare PoolFormer Model transformer outputting raw hidden-states without any specific head on top." , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: List[Any] = config
snake_case: Tuple = PoolFormerEncoder(SCREAMING_SNAKE_CASE__ )
# Initialize weights and apply final processing
self.post_init()
def _UpperCamelCase ( self ):
'''simple docstring'''
return self.embeddings.patch_embeddings
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Optional[int] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
snake_case: List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
if pixel_values is None:
raise ValueError('You have to specify pixel_values' )
snake_case: Optional[Any] = self.encoder(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: List[Any] = encoder_outputs[0]
if not return_dict:
return (sequence_output, None) + encoder_outputs[1:]
return BaseModelOutputWithNoAttention(
last_hidden_state=SCREAMING_SNAKE_CASE__ , hidden_states=encoder_outputs.hidden_states , )
class SCREAMING_SNAKE_CASE ( nn.Module ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__()
snake_case: Any = nn.Linear(config.hidden_size , config.hidden_size )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: int = self.dense(SCREAMING_SNAKE_CASE__ )
return output
@add_start_docstrings(
"\n PoolFormer Model transformer with an image classification head on top\n " , snake_case , )
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
def __init__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
super().__init__(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = config.num_labels
snake_case: str = PoolFormerModel(SCREAMING_SNAKE_CASE__ )
# Final norm
snake_case: int = PoolFormerGroupNorm(config.hidden_sizes[-1] )
# Classifier head
snake_case: Dict = (
nn.Linear(config.hidden_sizes[-1] , config.num_labels ) if config.num_labels > 0 else nn.Identity()
)
# Initialize weights and apply final processing
self.post_init()
@add_start_docstrings_to_model_forward(SCREAMING_SNAKE_CASE__ )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=SCREAMING_SNAKE_CASE__ , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = None , ):
'''simple docstring'''
snake_case: Union[str, Any] = return_dict if return_dict is not None else self.config.use_return_dict
snake_case: Optional[Any] = self.poolformer(
SCREAMING_SNAKE_CASE__ , output_hidden_states=SCREAMING_SNAKE_CASE__ , return_dict=SCREAMING_SNAKE_CASE__ , )
snake_case: Any = outputs[0]
snake_case: str = self.classifier(self.norm(SCREAMING_SNAKE_CASE__ ).mean([-2, -1] ) )
snake_case: Any = None
if labels is not None:
if self.config.problem_type is None:
if self.num_labels == 1:
snake_case: Tuple = 'regression'
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
snake_case: Dict = 'single_label_classification'
else:
snake_case: List[str] = 'multi_label_classification'
if self.config.problem_type == "regression":
snake_case: Union[str, Any] = MSELoss()
if self.num_labels == 1:
snake_case: List[str] = loss_fct(logits.squeeze() , labels.squeeze() )
else:
snake_case: int = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
elif self.config.problem_type == "single_label_classification":
snake_case: Union[str, Any] = CrossEntropyLoss()
snake_case: Dict = loss_fct(logits.view(-1 , self.num_labels ) , labels.view(-1 ) )
elif self.config.problem_type == "multi_label_classification":
snake_case: int = BCEWithLogitsLoss()
snake_case: Optional[int] = loss_fct(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )
if not return_dict:
snake_case: str = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return ImageClassifierOutputWithNoAttention(loss=SCREAMING_SNAKE_CASE__ , logits=SCREAMING_SNAKE_CASE__ , hidden_states=outputs.hidden_states ) | 692 | 0 |
import json
import os
from functools import lru_cache
from typing import List, Optional, Tuple
import regex as re
from ...tokenization_utils import AddedToken, PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {"vocab_file": "vocab.json", "merges_file": "merges.txt"}
# See all BART models at https://huggingface.co/models?filter=bart
__UpperCAmelCase = {
"vocab_file": {
"facebook/bart-base": "https://huggingface.co/facebook/bart-base/resolve/main/vocab.json",
"facebook/bart-large": "https://huggingface.co/facebook/bart-large/resolve/main/vocab.json",
"facebook/bart-large-mnli": "https://huggingface.co/facebook/bart-large-mnli/resolve/main/vocab.json",
"facebook/bart-large-cnn": "https://huggingface.co/facebook/bart-large-cnn/resolve/main/vocab.json",
"facebook/bart-large-xsum": "https://huggingface.co/facebook/bart-large-xsum/resolve/main/vocab.json",
"yjernite/bart_eli5": "https://huggingface.co/yjernite/bart_eli5/resolve/main/vocab.json",
},
"merges_file": {
"facebook/bart-base": "https://huggingface.co/facebook/bart-base/resolve/main/merges.txt",
"facebook/bart-large": "https://huggingface.co/facebook/bart-large/resolve/main/merges.txt",
"facebook/bart-large-mnli": "https://huggingface.co/facebook/bart-large-mnli/resolve/main/merges.txt",
"facebook/bart-large-cnn": "https://huggingface.co/facebook/bart-large-cnn/resolve/main/merges.txt",
"facebook/bart-large-xsum": "https://huggingface.co/facebook/bart-large-xsum/resolve/main/merges.txt",
"yjernite/bart_eli5": "https://huggingface.co/yjernite/bart_eli5/resolve/main/merges.txt",
},
}
__UpperCAmelCase = {
"facebook/bart-base": 1_024,
"facebook/bart-large": 1_024,
"facebook/bart-large-mnli": 1_024,
"facebook/bart-large-cnn": 1_024,
"facebook/bart-large-xsum": 1_024,
"yjernite/bart_eli5": 1_024,
}
@lru_cache()
def lowerCAmelCase_ ( ):
'''simple docstring'''
snake_case: Dict = (
list(range(ord('!' ) , ord('~' ) + 1 ) ) + list(range(ord('¡' ) , ord('¬' ) + 1 ) ) + list(range(ord('®' ) , ord('ÿ' ) + 1 ) )
)
snake_case: Optional[Any] = bs[:]
snake_case: Optional[int] = 0
for b in range(2**8 ):
if b not in bs:
bs.append(lowerCamelCase__ )
cs.append(2**8 + n )
n += 1
snake_case: Dict = [chr(lowerCamelCase__ ) for n in cs]
return dict(zip(lowerCamelCase__ , lowerCamelCase__ ) )
def lowerCAmelCase_ ( __A : Optional[int] ):
'''simple docstring'''
snake_case: Dict = set()
snake_case: Union[str, Any] = word[0]
for char in word[1:]:
pairs.add((prev_char, char) )
snake_case: Optional[Any] = char
return pairs
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="replace" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__="<mask>" , SCREAMING_SNAKE_CASE__=False , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: str = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else bos_token
snake_case: List[str] = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else eos_token
snake_case: Optional[int] = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else sep_token
snake_case: int = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else cls_token
snake_case: Optional[int] = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else unk_token
snake_case: Dict = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else pad_token
# Mask token behave like a normal word, i.e. include the space before it
snake_case: Union[str, Any] = AddedToken(UpperCamelCase_ , lstrip=UpperCamelCase_ , rstrip=UpperCamelCase_ ) if isinstance(UpperCamelCase_ , UpperCamelCase_ ) else mask_token
super().__init__(
errors=UpperCamelCase_ , bos_token=UpperCamelCase_ , eos_token=UpperCamelCase_ , unk_token=UpperCamelCase_ , sep_token=UpperCamelCase_ , cls_token=UpperCamelCase_ , pad_token=UpperCamelCase_ , mask_token=UpperCamelCase_ , add_prefix_space=UpperCamelCase_ , **UpperCamelCase_ , )
with open(UpperCamelCase_ , encoding='utf-8' ) as vocab_handle:
snake_case: int = json.load(UpperCamelCase_ )
snake_case: Any = {v: k for k, v in self.encoder.items()}
snake_case: Any = errors # how to handle errors in decoding
snake_case: str = bytes_to_unicode()
snake_case: List[str] = {v: k for k, v in self.byte_encoder.items()}
with open(UpperCamelCase_ , encoding='utf-8' ) as merges_handle:
snake_case: str = merges_handle.read().split('\n' )[1:-1]
snake_case: List[str] = [tuple(merge.split() ) for merge in bpe_merges]
snake_case: Union[str, Any] = dict(zip(UpperCamelCase_ , range(len(UpperCamelCase_ ) ) ) )
snake_case: Optional[int] = {}
snake_case: Optional[int] = add_prefix_space
# Should have added re.IGNORECASE so BPE merges can happen for capitalized versions of contractions
snake_case: Dict = re.compile(r'\'s|\'t|\'re|\'ve|\'m|\'ll|\'d| ?\p{L}+| ?\p{N}+| ?[^\s\p{L}\p{N}]+|\s+(?!\S)|\s+' )
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.encoder )
def _UpperCamelCase ( self ):
'''simple docstring'''
return dict(self.encoder , **self.added_tokens_encoder )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.cache:
return self.cache[token]
snake_case: List[str] = tuple(UpperCamelCase_ )
snake_case: str = get_pairs(UpperCamelCase_ )
if not pairs:
return token
while True:
snake_case: str = min(UpperCamelCase_ , key=lambda SCREAMING_SNAKE_CASE__ : self.bpe_ranks.get(UpperCamelCase_ , float('inf' ) ) )
if bigram not in self.bpe_ranks:
break
snake_case: List[Any] = bigram
snake_case: Any = []
snake_case: List[str] = 0
while i < len(UpperCamelCase_ ):
try:
snake_case: Union[str, Any] = word.index(UpperCamelCase_ , UpperCamelCase_ )
except ValueError:
new_word.extend(word[i:] )
break
else:
new_word.extend(word[i:j] )
snake_case: str = j
if word[i] == first and i < len(UpperCamelCase_ ) - 1 and word[i + 1] == second:
new_word.append(first + second )
i += 2
else:
new_word.append(word[i] )
i += 1
snake_case: Dict = tuple(UpperCamelCase_ )
snake_case: str = new_word
if len(UpperCamelCase_ ) == 1:
break
else:
snake_case: int = get_pairs(UpperCamelCase_ )
snake_case: Optional[int] = " ".join(UpperCamelCase_ )
snake_case: Dict = word
return word
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = []
for token in re.findall(self.pat , UpperCamelCase_ ):
snake_case: Any = "".join(
self.byte_encoder[b] for b in token.encode('utf-8' ) ) # Maps all our bytes to unicode strings, avoiding control tokens of the BPE (spaces in our case)
bpe_tokens.extend(bpe_token for bpe_token in self.bpe(UpperCamelCase_ ).split(' ' ) )
return bpe_tokens
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.encoder.get(UpperCamelCase_ , self.encoder.get(self.unk_token ) )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.decoder.get(UpperCamelCase_ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = "".join(UpperCamelCase_ )
snake_case: Union[str, Any] = bytearray([self.byte_decoder[c] for c in text] ).decode('utf-8' , errors=self.errors )
return text
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(UpperCamelCase_ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: Any = os.path.join(
UpperCamelCase_ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
snake_case: Optional[int] = os.path.join(
UpperCamelCase_ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['merges_file'] )
with open(UpperCamelCase_ , 'w' , encoding='utf-8' ) as f:
f.write(json.dumps(self.encoder , indent=2 , sort_keys=UpperCamelCase_ , ensure_ascii=UpperCamelCase_ ) + '\n' )
snake_case: str = 0
with open(UpperCamelCase_ , 'w' , encoding='utf-8' ) as writer:
writer.write('#version: 0.2\n' )
for bpe_tokens, token_index in sorted(self.bpe_ranks.items() , key=lambda SCREAMING_SNAKE_CASE__ : kv[1] ):
if index != token_index:
logger.warning(
F"""Saving vocabulary to {merge_file}: BPE merge indices are not consecutive."""
' Please check that the tokenizer is not corrupted!' )
snake_case: str = token_index
writer.write(' '.join(UpperCamelCase_ ) + '\n' )
index += 1
return vocab_file, merge_file
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.cls_token_id] + token_ids_a + [self.sep_token_id]
snake_case: List[Any] = [self.cls_token_id]
snake_case: Tuple = [self.sep_token_id]
return cls + token_ids_a + sep + sep + token_ids_a + sep
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCamelCase_ , token_ids_a=UpperCamelCase_ , already_has_special_tokens=UpperCamelCase_ )
if token_ids_a is None:
return [1] + ([0] * len(UpperCamelCase_ )) + [1]
return [1] + ([0] * len(UpperCamelCase_ )) + [1, 1] + ([0] * len(UpperCamelCase_ )) + [1]
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
snake_case: List[str] = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep + sep + token_ids_a + sep ) * [0]
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__=False , **SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: List[str] = kwargs.pop('add_prefix_space' , self.add_prefix_space )
if (is_split_into_words or add_prefix_space) and (len(UpperCamelCase_ ) > 0 and not text[0].isspace()):
snake_case: Tuple = " " + text
return (text, kwargs) | 710 |
'''simple docstring'''
from queue import PriorityQueue
from typing import Any
import numpy as np
def lowerCAmelCase_ ( __A : dict , __A : str , __A : set , __A : set , __A : dict , __A : dict , __A : PriorityQueue , __A : dict , __A : float | int , ):
'''simple docstring'''
for nxt, d in graph[v]:
if nxt in visited_forward:
continue
snake_case: Any = cst_fwd.get(__A , np.inf )
snake_case: int = cst_fwd[v] + d
if new_cost_f < old_cost_f:
queue.put((new_cost_f, nxt) )
snake_case: Union[str, Any] = new_cost_f
snake_case: Tuple = v
if nxt in visited_backward:
if cst_fwd[v] + d + cst_bwd[nxt] < shortest_distance:
snake_case: List[str] = cst_fwd[v] + d + cst_bwd[nxt]
return shortest_distance
def lowerCAmelCase_ ( __A : str , __A : str , __A : dict , __A : dict ):
'''simple docstring'''
snake_case: Optional[Any] = -1
snake_case: Any = set()
snake_case: str = set()
snake_case: int = {source: 0}
snake_case: Dict = {destination: 0}
snake_case: int = {source: None}
snake_case: Union[str, Any] = {destination: None}
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: PriorityQueue[Any] = PriorityQueue()
snake_case: Tuple = np.inf
queue_forward.put((0, source) )
queue_backward.put((0, destination) )
if source == destination:
return 0
while not queue_forward.empty() and not queue_backward.empty():
snake_case , snake_case: List[str] = queue_forward.get()
visited_forward.add(__A )
snake_case , snake_case: int = queue_backward.get()
visited_backward.add(__A )
snake_case: str = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
snake_case: Optional[Any] = pass_and_relaxation(
__A , __A , __A , __A , __A , __A , __A , __A , __A , )
if cst_fwd[v_fwd] + cst_bwd[v_bwd] >= shortest_distance:
break
if shortest_distance != np.inf:
snake_case: Any = shortest_distance
return shortest_path_distance
__UpperCAmelCase = {
"B": [["C", 1]],
"C": [["D", 1]],
"D": [["F", 1]],
"E": [["B", 1], ["G", 2]],
"F": [],
"G": [["F", 1]],
}
__UpperCAmelCase = {
"B": [["E", 1]],
"C": [["B", 1]],
"D": [["C", 1]],
"F": [["D", 1], ["G", 1]],
"E": [[None, np.inf]],
"G": [["E", 2]],
}
if __name__ == "__main__":
import doctest
doctest.testmod() | 692 | 0 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import List, Optional, Tuple
from ...tokenization_utils import AddedToken
from ...tokenization_utils_fast import PreTrainedTokenizerFast
from ...utils import is_sentencepiece_available, logging
if is_sentencepiece_available():
from .tokenization_albert import AlbertTokenizer
else:
__UpperCAmelCase = None
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = {'vocab_file': 'spiece.model', 'tokenizer_file': 'tokenizer.json'}
__UpperCAmelCase = {
'vocab_file': {
'albert-base-v1': 'https://huggingface.co/albert-base-v1/resolve/main/spiece.model',
'albert-large-v1': 'https://huggingface.co/albert-large-v1/resolve/main/spiece.model',
'albert-xlarge-v1': 'https://huggingface.co/albert-xlarge-v1/resolve/main/spiece.model',
'albert-xxlarge-v1': 'https://huggingface.co/albert-xxlarge-v1/resolve/main/spiece.model',
'albert-base-v2': 'https://huggingface.co/albert-base-v2/resolve/main/spiece.model',
'albert-large-v2': 'https://huggingface.co/albert-large-v2/resolve/main/spiece.model',
'albert-xlarge-v2': 'https://huggingface.co/albert-xlarge-v2/resolve/main/spiece.model',
'albert-xxlarge-v2': 'https://huggingface.co/albert-xxlarge-v2/resolve/main/spiece.model',
},
'tokenizer_file': {
'albert-base-v1': 'https://huggingface.co/albert-base-v1/resolve/main/tokenizer.json',
'albert-large-v1': 'https://huggingface.co/albert-large-v1/resolve/main/tokenizer.json',
'albert-xlarge-v1': 'https://huggingface.co/albert-xlarge-v1/resolve/main/tokenizer.json',
'albert-xxlarge-v1': 'https://huggingface.co/albert-xxlarge-v1/resolve/main/tokenizer.json',
'albert-base-v2': 'https://huggingface.co/albert-base-v2/resolve/main/tokenizer.json',
'albert-large-v2': 'https://huggingface.co/albert-large-v2/resolve/main/tokenizer.json',
'albert-xlarge-v2': 'https://huggingface.co/albert-xlarge-v2/resolve/main/tokenizer.json',
'albert-xxlarge-v2': 'https://huggingface.co/albert-xxlarge-v2/resolve/main/tokenizer.json',
},
}
__UpperCAmelCase = {
'albert-base-v1': 512,
'albert-large-v1': 512,
'albert-xlarge-v1': 512,
'albert-xxlarge-v1': 512,
'albert-base-v2': 512,
'albert-large-v2': 512,
'albert-xlarge-v2': 512,
'albert-xxlarge-v2': 512,
}
__UpperCAmelCase = '▁'
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = AlbertTokenizer
def __init__( self , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=None , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=True , SCREAMING_SNAKE_CASE__=False , SCREAMING_SNAKE_CASE__="[CLS]" , SCREAMING_SNAKE_CASE__="[SEP]" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="[SEP]" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__="[CLS]" , SCREAMING_SNAKE_CASE__="[MASK]" , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Union[str, Any] = (
AddedToken(__A , lstrip=__A , rstrip=__A , normalized=__A )
if isinstance(__A , __A )
else mask_token
)
super().__init__(
__A , tokenizer_file=__A , do_lower_case=__A , remove_space=__A , keep_accents=__A , bos_token=__A , eos_token=__A , unk_token=__A , sep_token=__A , pad_token=__A , cls_token=__A , mask_token=__A , **__A , )
snake_case: Dict = do_lower_case
snake_case: List[str] = remove_space
snake_case: Any = keep_accents
snake_case: Tuple = vocab_file
snake_case: Optional[int] = False if not self.vocab_file else True
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: List[Any] = [self.sep_token_id]
snake_case: Any = [self.cls_token_id]
if token_ids_a is None:
return cls + token_ids_a + sep
return cls + token_ids_a + sep + token_ids_a + sep
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: Optional[Any] = [self.sep_token_id]
snake_case: Union[str, Any] = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1]
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not self.can_save_slow_tokenizer:
raise ValueError(
'Your fast tokenizer does not have the necessary information to save the vocabulary for a slow '
'tokenizer.' )
if not os.path.isdir(__A ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: Union[str, Any] = os.path.join(
__A , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(__A ):
copyfile(self.vocab_file , __A )
return (out_vocab_file,) | 711 |
'''simple docstring'''
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__UpperCAmelCase = logging.get_logger(__name__)
__UpperCAmelCase = "▁"
__UpperCAmelCase = {"vocab_file": "sentencepiece.bpe.model"}
__UpperCAmelCase = {
"vocab_file": {
"facebook/xglm-564M": "https://huggingface.co/facebook/xglm-564M/resolve/main/sentencepiece.bpe.model",
}
}
__UpperCAmelCase = {
"facebook/xglm-564M": 2_048,
}
class SCREAMING_SNAKE_CASE ( snake_case ):
'''simple docstring'''
__UpperCamelCase = VOCAB_FILES_NAMES
__UpperCamelCase = PRETRAINED_VOCAB_FILES_MAP
__UpperCamelCase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCamelCase = ["input_ids", "attention_mask"]
def __init__( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="</s>" , SCREAMING_SNAKE_CASE__="<s>" , SCREAMING_SNAKE_CASE__="<unk>" , SCREAMING_SNAKE_CASE__="<pad>" , SCREAMING_SNAKE_CASE__ = None , **SCREAMING_SNAKE_CASE__ , ):
'''simple docstring'''
snake_case: Optional[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
# Compatibility with the original tokenizer
snake_case: Optional[Any] = 7
snake_case: List[str] = [F"""<madeupword{i}>""" for i in range(self.num_madeup_words )]
snake_case: str = kwargs.get('additional_special_tokens' , [] )
kwargs["additional_special_tokens"] += [
word for word in madeup_words if word not in kwargs["additional_special_tokens"]
]
super().__init__(
bos_token=SCREAMING_SNAKE_CASE__ , eos_token=SCREAMING_SNAKE_CASE__ , unk_token=SCREAMING_SNAKE_CASE__ , sep_token=SCREAMING_SNAKE_CASE__ , cls_token=SCREAMING_SNAKE_CASE__ , pad_token=SCREAMING_SNAKE_CASE__ , sp_model_kwargs=self.sp_model_kwargs , **SCREAMING_SNAKE_CASE__ , )
snake_case: int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(SCREAMING_SNAKE_CASE__ ) )
snake_case: int = vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# The first "real" token "," has position 4 in the original fairseq vocab and position 3 in the spm vocab
snake_case: Tuple = 1
# Mimic fairseq token-to-id alignment for the first 4 token
snake_case: Optional[Any] = {'<s>': 0, '<pad>': 1, '</s>': 2, '<unk>': 3}
snake_case: Union[str, Any] = len(self.sp_model )
snake_case: str = {F"""<madeupword{i}>""": sp_size + i + self.fairseq_offset for i in range(self.num_madeup_words )}
self.fairseq_tokens_to_ids.update(SCREAMING_SNAKE_CASE__ )
snake_case: Union[str, Any] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def __getstate__( self ):
'''simple docstring'''
snake_case: List[Any] = self.__dict__.copy()
snake_case: Union[str, Any] = None
snake_case: Union[str, Any] = self.sp_model.serialized_model_proto()
return state
def __setstate__( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[int] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
snake_case: Union[str, Any] = {}
snake_case: Tuple = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.LoadFromSerializedProto(self.sp_model_proto )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if token_ids_a is None:
return [self.sep_token_id] + token_ids_a
snake_case: Optional[Any] = [self.sep_token_id]
return sep + token_ids_a + sep + sep + token_ids_a
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None , SCREAMING_SNAKE_CASE__ = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=SCREAMING_SNAKE_CASE__ , token_ids_a=SCREAMING_SNAKE_CASE__ , already_has_special_tokens=SCREAMING_SNAKE_CASE__ )
if token_ids_a is None:
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
return [1] + ([0] * len(SCREAMING_SNAKE_CASE__ )) + [1, 1] + ([0] * len(SCREAMING_SNAKE_CASE__ ))
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
snake_case: int = [self.sep_token_id]
if token_ids_a is None:
return len(sep + token_ids_a ) * [0]
return len(sep + token_ids_a + sep + sep + token_ids_a ) * [0]
@property
def _UpperCamelCase ( self ):
'''simple docstring'''
return len(self.sp_model ) + self.fairseq_offset + self.num_madeup_words
def _UpperCamelCase ( self ):
'''simple docstring'''
snake_case: Optional[int] = {self.convert_ids_to_tokens(SCREAMING_SNAKE_CASE__ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
return self.sp_model.encode(SCREAMING_SNAKE_CASE__ , out_type=SCREAMING_SNAKE_CASE__ )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
snake_case: Dict = self.sp_model.PieceToId(SCREAMING_SNAKE_CASE__ )
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset )
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ ):
'''simple docstring'''
snake_case: Optional[Any] = ''.join(SCREAMING_SNAKE_CASE__ ).replace(SCREAMING_SNAKE_CASE__ , ' ' ).strip()
return out_string
def _UpperCamelCase ( self , SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = None ):
'''simple docstring'''
if not os.path.isdir(SCREAMING_SNAKE_CASE__ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case: List[str] = os.path.join(
SCREAMING_SNAKE_CASE__ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(SCREAMING_SNAKE_CASE__ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , SCREAMING_SNAKE_CASE__ )
elif not os.path.isfile(self.vocab_file ):
with open(SCREAMING_SNAKE_CASE__ , 'wb' ) as fi:
snake_case: int = self.sp_model.serialized_model_proto()
fi.write(SCREAMING_SNAKE_CASE__ )
return (out_vocab_file,) | 692 | 0 |
'''simple docstring'''
def lowerCAmelCase_ ( __A : List[str] , __A : Tuple ):
'''simple docstring'''
return "\n".join(
f"""{number} * {i} = {number * i}""" for i in range(1 , number_of_terms + 1 ) )
if __name__ == "__main__":
print(multiplication_table(number=5, number_of_terms=10)) | 712 |
'''simple docstring'''
from operator import delitem, getitem, setitem
import pytest
from data_structures.hashing.hash_map import HashMap
def lowerCAmelCase_ ( __A : Optional[Any] ):
'''simple docstring'''
return getitem, k
def lowerCAmelCase_ ( __A : Any , __A : Optional[int] ):
'''simple docstring'''
return setitem, k, v
def lowerCAmelCase_ ( __A : List[str] ):
'''simple docstring'''
return delitem, k
def lowerCAmelCase_ ( __A : str , __A : int , *__A : Tuple ):
'''simple docstring'''
try:
return fun(__A , *__A ), None
except Exception as e:
return None, e
__UpperCAmelCase = (
_set("key_a", "val_a"),
_set("key_b", "val_b"),
)
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_a", "val_b"),
]
__UpperCAmelCase = [
_set("key_a", "val_a"),
_set("key_b", "val_b"),
_del("key_a"),
_del("key_b"),
_set("key_a", "val_a"),
_del("key_a"),
]
__UpperCAmelCase = [
_get("key_a"),
_del("key_a"),
_set("key_a", "val_a"),
_del("key_a"),
_del("key_a"),
_get("key_a"),
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
]
__UpperCAmelCase = [
*[_set(x, x) for x in range(5)], # guaranteed upsize
*[_del(x) for x in range(5)],
_set("key_a", "val_b"),
]
@pytest.mark.parametrize(
'operations' , (
pytest.param(_add_items , id='add items' ),
pytest.param(_overwrite_items , id='overwrite items' ),
pytest.param(_delete_items , id='delete items' ),
pytest.param(_access_absent_items , id='access absent items' ),
pytest.param(_add_with_resize_up , id='add with resize up' ),
pytest.param(_add_with_resize_down , id='add with resize down' ),
) , )
def lowerCAmelCase_ ( __A : str ):
'''simple docstring'''
snake_case: List[Any] = HashMap(initial_block_size=4 )
snake_case: List[Any] = {}
for _, (fun, *args) in enumerate(__A ):
snake_case , snake_case: Optional[int] = _run_operation(__A , __A , *__A )
snake_case , snake_case: str = _run_operation(__A , __A , *__A )
assert my_res == py_res
assert str(__A ) == str(__A )
assert set(__A ) == set(__A )
assert len(__A ) == len(__A )
assert set(my.items() ) == set(py.items() )
def lowerCAmelCase_ ( ):
'''simple docstring'''
def is_public(__A : str ) -> bool:
return not name.startswith('_' )
snake_case: Dict = {name for name in dir({} ) if is_public(__A )}
snake_case: List[str] = {name for name in dir(HashMap() ) if is_public(__A )}
assert dict_public_names > hash_public_names | 692 | 0 |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.