Update app.py

app.py

@@ -1,6 +1,3 @@
-import os
-import asyncio
-from concurrent.futures import ThreadPoolExecutor
 from fastapi import FastAPI, File, UploadFile
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.middleware.gzip import GZipMiddleware
@@ -19,21 +16,15 @@ app.add_middleware(
     allow_methods=["*"],
     allow_headers=["*"]
 )
-app.add_middleware(GZipMiddleware, minimum_size=1000)
 
 # Initialize models once at startup
 ocr_model = ocr_predictor(pretrained=True)
-paddle_ocr = PaddleOCR(lang='en', use_angle_cls=True, use_gpu=True)
-
-# Get the number of available CPUs
-num_cpus = os.cpu_count()
-
-# Initialize ThreadPoolExecutor with 2 workers
-executor = ThreadPoolExecutor(max_workers=num_cpus)
+paddle_ocr = PaddleOCR(lang='en', use_angle_cls=True)
 
 def ocr_with_doctr(file):
     text_output = ''
     doc = DocumentFile.from_pdf(file)
+    ocr_model = ocr_predictor(pretrained=True)
     result = ocr_model(doc)
     for page in result.pages:
         for block in page.blocks:
@@ -52,20 +43,16 @@ def ocr_with_paddle(img):
 def generate_text_from_image(img):
     return ocr_with_paddle(img)
 
-async def run_blocking_func(func, *args):
-    loop = asyncio.get_event_loop()
-    return await loop.run_in_executor(executor, func, *args)
-
 @app.post("/ocr/")
 async def perform_ocr(file: UploadFile = File(...)):
     file_bytes = await file.read()
     if file.filename.endswith('.pdf'):
-        text_output = await run_blocking_func(ocr_with_doctr, io.BytesIO(file_bytes))
+        text_output = ocr_with_doctr(io.BytesIO(file_bytes))
     else:
         img = np.array(Image.open(io.BytesIO(file_bytes)))
-        text_output = await run_blocking_func(generate_text_from_image, img)
+        text_output = generate_text_from_image(img)
     return {"ocr_text": text_output}
 
 @app.get("/test/")
 async def test_call():
-    return {"message": "Hi. I'm running"}
+    return {"message": "Hi. I'm running"}