Spaces:
Running
Running
File size: 2,608 Bytes
c965e8a 08b129e c965e8a 14f8a5c c965e8a 14f8a5c c965e8a 14f8a5c c965e8a 14f8a5c c965e8a 14f8a5c c965e8a 14f8a5c c965e8a 5c9215b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
from knowledgeassistant.exception.exception import KnowledgeAssistantException
from knowledgeassistant.logging.logger import logging
from knowledgeassistant.entity.config_entity import DataSummarizationConfig
from knowledgeassistant.utils.main_utils.utils import write_txt_file, read_txt_file
import sys
import torch
from transformers import pipeline, AutoTokenizer
class DataSummarization:
def __init__(self, data_summarization_config: DataSummarizationConfig):
try:
self.data_summarization_config = data_summarization_config
except Exception as e:
raise KnowledgeAssistantException(e, sys)
def summarize(self, input_text_path: str, min_length: int):
try:
model_path = "/app/models/bart-large-cnn"
tokenizer = AutoTokenizer.from_pretrained(model_path)
pipe = pipeline("summarization", model=model_path, tokenizer=model_path)
logging.info("Summarization Pipeline Successfully Setup")
text = read_txt_file(input_text_path)
tokens = tokenizer.encode(text, truncation=True, max_length=1024, return_tensors="pt")
if len(tokens[0]) >= 1024:
logging.warning("Input text exceeded 1024 tokens. It has been truncated.")
truncated_text = tokenizer.decode(tokens[0], skip_special_tokens=True)
frontend_message = "Your input text exceeded the limit of 1024 tokens and has been truncated."
else:
truncated_text = text
frontend_message = ""
# Generate summary
summary = pipe(truncated_text, min_length=min_length, max_length=142, do_sample=False)
logging.info("Text successfully summarized")
# Save summary
write_txt_file(self.data_summarization_config.summarized_text_file_path, summary[0].get("summary_text"))
logging.info("Successfully wrote summarized text")
# Return summary along with frontend message
return {
"summary": summary[0].get("summary_text"),
"warning": frontend_message
}
except Exception as e:
raise KnowledgeAssistantException(e, sys)
def initiate_data_summarization(self, input_text_path: str, min_length: int):
try:
self.summarize(
input_text_path = input_text_path,
min_length = min_length
)
except Exception as e:
raise KnowledgeAssistantException(e, sys) |