Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,16 +1,39 @@
|
|
1 |
import gradio as gr
|
2 |
import torch
|
3 |
-
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
4 |
import re
|
|
|
|
|
5 |
|
6 |
-
#
|
7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
tokenizer = AutoTokenizer.from_pretrained("pepegiallo/flan-t5-base_ner")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
model.eval()
|
10 |
|
|
|
11 |
id2label = {0: "LOC", 1: "ORG", 2: "PER", 3: "O"}
|
12 |
|
13 |
-
#
|
14 |
def custom_tokenize(text):
|
15 |
return re.findall(r"\w+|[^\w\s]", text, re.UNICODE)
|
16 |
|
@@ -22,6 +45,7 @@ def custom_detokenize(tokens):
|
|
22 |
text += token
|
23 |
return text
|
24 |
|
|
|
25 |
def classify_tokens(text):
|
26 |
tokens = custom_tokenize(text)
|
27 |
results = []
|
@@ -32,14 +56,14 @@ def classify_tokens(text):
|
|
32 |
|
33 |
inputs = tokenizer(prompt, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
|
34 |
with torch.no_grad():
|
35 |
-
logits = model(**inputs)
|
36 |
pred_id = torch.argmax(logits, dim=-1).item()
|
37 |
label = id2label[pred_id]
|
38 |
|
39 |
results.append((tokens[i], label))
|
40 |
return results
|
41 |
|
42 |
-
# Gradio
|
43 |
demo = gr.Interface(
|
44 |
fn=classify_tokens,
|
45 |
inputs=gr.Textbox(lines=3, placeholder="Enter a sentence..."),
|
|
|
1 |
import gradio as gr
|
2 |
import torch
|
|
|
3 |
import re
|
4 |
+
from transformers import AutoTokenizer, T5EncoderModel
|
5 |
+
import torch.nn as nn
|
6 |
|
7 |
+
# Klassendefinition aus dem Training
|
8 |
+
class FlanT5Classifier(nn.Module):
|
9 |
+
def __init__(self, base_model_name="google/flan-t5-base", num_labels=4):
|
10 |
+
super().__init__()
|
11 |
+
self.encoder = T5EncoderModel.from_pretrained(base_model_name)
|
12 |
+
self.dropout = nn.Dropout(0.1)
|
13 |
+
self.classifier = nn.Linear(self.encoder.config.d_model, num_labels)
|
14 |
+
|
15 |
+
def forward(self, input_ids, attention_mask=None):
|
16 |
+
encoder_outputs = self.encoder(input_ids=input_ids, attention_mask=attention_mask)
|
17 |
+
pooled = encoder_outputs.last_hidden_state[:, 0]
|
18 |
+
logits = self.classifier(self.dropout(pooled))
|
19 |
+
return {"logits": logits}
|
20 |
+
|
21 |
+
# Tokenizer laden
|
22 |
tokenizer = AutoTokenizer.from_pretrained("pepegiallo/flan-t5-base_ner")
|
23 |
+
|
24 |
+
# Modell instanziieren und Token-Embeddings anpassen
|
25 |
+
model = FlanT5Classifier()
|
26 |
+
model.encoder.resize_token_embeddings(len(tokenizer))
|
27 |
+
|
28 |
+
# Gewichte laden
|
29 |
+
state_dict = torch.load("pytorch_model.bin", map_location="cpu")
|
30 |
+
model.load_state_dict(state_dict)
|
31 |
model.eval()
|
32 |
|
33 |
+
# ID-Zuordnung
|
34 |
id2label = {0: "LOC", 1: "ORG", 2: "PER", 3: "O"}
|
35 |
|
36 |
+
# Tokenizer-Funktionen
|
37 |
def custom_tokenize(text):
|
38 |
return re.findall(r"\w+|[^\w\s]", text, re.UNICODE)
|
39 |
|
|
|
45 |
text += token
|
46 |
return text
|
47 |
|
48 |
+
# Klassifikationsfunktion
|
49 |
def classify_tokens(text):
|
50 |
tokens = custom_tokenize(text)
|
51 |
results = []
|
|
|
56 |
|
57 |
inputs = tokenizer(prompt, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
|
58 |
with torch.no_grad():
|
59 |
+
logits = model(**inputs)["logits"]
|
60 |
pred_id = torch.argmax(logits, dim=-1).item()
|
61 |
label = id2label[pred_id]
|
62 |
|
63 |
results.append((tokens[i], label))
|
64 |
return results
|
65 |
|
66 |
+
# Gradio UI
|
67 |
demo = gr.Interface(
|
68 |
fn=classify_tokens,
|
69 |
inputs=gr.Textbox(lines=3, placeholder="Enter a sentence..."),
|