change

app.py

@@ -4,10 +4,11 @@ from fastapi import FastAPI,HTTPException,status,UploadFile,File
 from pydantic import BaseModel
 import uvicorn
 import logging
-from PIL import Image
+from PIL import Image,UnidentifiedImageError
 import pytesseract
 from io import BytesIO
 import os
+import re
 
 # Configurer les répertoires de cache
 os.environ['TRANSFORMERS_CACHE'] = '/app/.cache'
@@ -32,8 +33,7 @@ app =FastAPI(
 logging.basicConfig(level=logging.INFO)
 logger =logging.getLogger(__name__)
 summarize =pipeline('summarization', model="facebook/bart-large-cnn")
-generated_text = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large") 
-# translation_pieline =pipeline("translation_en_to_fr",model="Helsinki-NLP/opus-mt-en-fr")
+pipe = pipeline("summarization", model="plguillou/t5-base-fr-sum-cnndm") 
 classify_zero_shot = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
 
 class TextSummary(BaseModel):
@@ -50,7 +50,9 @@ async def home():
 async def summary_text_bart(input:TextSummary):
     "add text to summarize"
     try:
-        summary = summarize(input.text,do_sample=False)
+        
+        preprocessed_text_summary =re.sub(r'\s+',' ',input.text).strip()
+        summary = summarize(preprocessed_text_summary,do_sample=False)
         summary_text =summary[0].get('summary_text')
         result = classify_zero_shot(
                 summary_text,
@@ -66,7 +68,7 @@ async def summary_text_bart(input:TextSummary):
         return {
             "summary_text" :summary_text,
             "Statut":formatted_result,
-            # "format":formatted_result,
+            # "preprocessed_text_summary":preprocessed_text_summary,
             "len_input" : len(input.text),
             "len_output" :len(summary_text)
 
@@ -84,17 +86,30 @@ async def summary_text_bart(input:TextSummary):
 async def upload_image(image: UploadFile = File(...)):
     contents = await image.read()
     try:
-        image_open = Image.open(BytesIO(contents))
+
+        try:
+            image_open = Image.open(BytesIO(contents))
+        except UnidentifiedImageError:
+            raise HTTPException(status_code=400, detail="Unsupported or corrupted image file.")
+ 
         raw_text = pytesseract.image_to_string(image_open,lang='fra')
-        summary = summarize(raw_text,do_sample=False)
-        summary_text =summary[0].get('summary_text')
+        logger.info(f"Extract text:{raw_text}")
+        #preprocessing the text
+        preprocessed_text =re.sub(r'\s+',' ',raw_text).strip()
+        logger.info(f"Preprocessing  text:{preprocessed_text}")
+        text_summary = pipe(raw_text,do_sample=False)
+        summary_text_image =text_summary[0].get('summary_text')
         return {
-                # "text": raw_text,
-                "summary":summary_text}
+                # "text": preprocessed_text,
+                # "len":len(preprocessed_text),
+                "summary":summary_text_image,
+                "len summary":len(summary_text_image)
+                }
     except Exception as e:
-        return {"error": str(e)}
+        logger.error(f"Error processing image:{e}")
+        raise HTTPException(status_code=500, detail=str(e))
 if __name__ == "__main__":
-    uvicorn.run("main:app",host="0.0.0.0",port=8000,reload=True)
+    uvicorn.run("app:app",host="0.0.0.0",port=8000,reload=True)