Update app.py

app.py

@@ -1,144 +1,29 @@
-import os
-import base64
 import gradio as gr
-import logging
-from huggingface_hub import InferenceClient
+from transformers import TrOCRProcessor, VisionEncoderDecoderModel
 from PIL import Image
-import io
-
-# Configure logging
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
-
-# Optional imports with error handling
-try:
-    from pdf2image import convert_from_path
-    import easyocr
-    HAS_OCR = True
-except ImportError:
-    HAS_OCR = False
-    print("OCR features will be disabled. Install pdf2image and easyocr for full functionality.")
-
-# Initialize Hugging Face client
-client = InferenceClient(model="google/gemma-7b-it")
-
-def initialize_ocr():
-    """Initialize OCR with error handling"""
-    if not HAS_OCR:
-        return None
-    try:
-        return easyocr.Reader(['en'], gpu=True)  # Enable GPU if available
-    except Exception as e:
-        logger.error(f"Failed to initialize OCR: {e}")
-        return None
-
-reader = initialize_ocr()
-
-# Add custom CSS for better styling
-custom_css = """
-    .container { max-width: 1200px; margin: auto; }
-    .gradio-container { font-family: 'Arial', sans-serif; }
-    .gr-button { background-color: #2196F3 !important; color: white !important; }
-    .gr-button:hover { background-color: #1976D2 !important; }
-    .feedback { margin-top: 20px; padding: 10px; border-radius: 4px; }
-    .success { background-color: #4CAF50; color: white; }
-    .error { background-color: #f44336; color: white; }
-    .footer { text-align: center; margin-top: 20px; color: #666; }
-"""
-
-def make_api_call(text_content, image_path=None, retries=3):
-    """Enhanced API call handler with better prompting"""
-    try:
-        prompt = f"""Analyze this document and provide a detailed analysis with:
-
-        📝 CORRECTED TEXT:
-        [Provide the text with proper formatting, corrected spelling and grammar]
-
-        📋 SUMMARY:
-        [A concise 2-3 sentence summary of the main content]
-
-        🔑 KEY POINTS:
-        [List the 3-5 most important points]
-
-        🏷️ ENTITIES DETECTED:
-        - Dates: [List any dates found]
-        - Names: [List any names found]
-        - Numbers/Values: [List any significant numbers/values]
-        - Organizations: [List any organizations mentioned]
-
-        📄 Original Content: {text_content}"""
-
-        for attempt in range(retries):
-            try:
-                response = client.text_generation(
-                    prompt=prompt,
-                    max_new_tokens=1500,  # Increased token limit
-                    temperature=0.7,
-                    top_p=0.95,
-                )
-                return response
-            except Exception as e:
-                if attempt == retries - 1:
-                    raise e
-                logger.warning(f"Attempt {attempt + 1}/{retries} failed, retrying...")
-                continue
-
-    except Exception as e:
-        logger.error(f"API call failed: {e}")
-        return f"Error processing request: {str(e)}"
-
-# ... rest of your existing code ...
-
-# Create enhanced Gradio interface
-demo = gr.Interface(
-    fn=process_and_analyze,
-    inputs=[
-        gr.File(
-            label="📎 Upload Document",
-            file_types=[".pdf", ".png", ".jpg", ".jpeg"],
-            type="file"
-        )
-    ],
-    outputs=[
-        gr.Textbox(
-            label="🔍 Analysis Results",
-            lines=20,
-            show_copy_button=True
-        )
-    ],
-    title="🤖 Smart Document Analyzer Pro",
-    description="""
-    ### Upload your documents for instant AI-powered analysis!
-    
-    This tool can:
-    - 📝 Extract and correct text from images and PDFs
-    - 📊 Provide detailed summaries and key points
-    - 🔍 Identify important entities (dates, names, numbers)
-    - ✨ Format and structure the content
-    """,
-    examples=[
-        ["example1.pdf"],
-        ["example2.jpg"],
-    ],
-    theme=gr.themes.Soft().set(
-        primary_hue="blue",
-        secondary_hue="indigo",
-    ),
-    css=custom_css,
-    allow_flagging="never",
+import requests
+from io import BytesIO
+
+# Load TrOCR model and processor
+processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-handwritten")
+model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-handwritten")
+
+def extract_text_from_image(image):
+    """Extract text from an uploaded image using Hugging Face TrOCR model."""
+    image = image.convert("RGB")
+    pixel_values = processor(image, return_tensors="pt").pixel_values
+    generated_ids = model.generate(pixel_values)
+    extracted_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    return extracted_text
+
+# Create Gradio Interface
+interface = gr.Interface(
+    fn=extract_text_from_image,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Textbox(label="Extracted Text"),
+    title="OCR Text Extractor",
+    description="Upload an image to extract text using Hugging Face's TrOCR model."
 )
 
-# Add markdown for footer
-demo.footer = """
-<div class="footer">
-    <p>🚀 Powered by Hugging Face & EasyOCR | Built with Gradio</p>
-    <p>For optimal results, use clear images or well-scanned PDFs</p>
-</div>
-"""
-
 if __name__ == "__main__":
-    demo.launch(
-        share=True,
-        enable_queue=True,
-        show_error=True,
-    )
+    interface.launch(share=True)