Update app.py

app.py

@@ -1,68 +1,58 @@
-from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor
+# app.py
+
 import gradio as gr
 from PIL import Image
-
-# Use a public model identifier. If you need a private model, remember to authenticate.
-model_name = "google/pix2struct-textcaps-base"
-model = Pix2StructForConditionalGeneration.from_pretrained(model_name)
-processor = Pix2StructProcessor.from_pretrained(model_name)
+import pytesseract
+import sympy
 
 def solve_math_problem(image):
     try:
-        # Ensure the image is in RGB format.
-        image = image.convert("RGB")
+        # Convert image to grayscale for better OCR performance
+        image = image.convert("L")
+        
+        # Preprocess the image (optional enhancements can be added here)
+        # For example, image = image.point(lambda x: 0 if x < 140 else 255, '1')
+        
+        # Use pytesseract to extract text from the image
+        problem_text = pytesseract.image_to_string(image, config='--psm 7')
         
-        # Preprocess the image and text. Note that header_text is omitted as it's not used for non-VQA tasks.
-        inputs = processor(
-            images=[image],
-            text="Solve the following math problem:",
-            return_tensors="pt",
-            max_patches=2048
-        )
+        # Clean and prepare the extracted text
+        problem_text = problem_text.strip().replace('\n', '').replace(' ', '')
         
-        # Generate the solution with generation parameters.
-        predictions = model.generate(
-            **inputs,
-            max_new_tokens=200,
-            early_stopping=True,
-            num_beams=4,
-            temperature=0.2
-        )
+        # Use sympy to parse and solve the equation
+        # Handle simple arithmetic and algebraic equations
+        expr = sympy.sympify(problem_text)
+        solution = sympy.solve(expr)
         
-        # Decode the problem text and generated solution.
-        problem_text = processor.decode(
-            inputs["input_ids"][0],
-            skip_special_tokens=True,
-            clean_up_tokenization_spaces=True
-        )
-        solution = processor.decode(
-            predictions[0],
-            skip_special_tokens=True,
-            clean_up_tokenization_spaces=True
-        )
+        # Format the solution for display
+        if isinstance(solution, list):
+            solution = ', '.join([str(s) for s in solution])
+        else:
+            solution = str(solution)
         
-        return f"Problem: {problem_text}\nSolution: {solution}"
+        return f"**Problem:** {problem_text}\n\n**Solution:** {solution}"
         
     except Exception as e:
-        return f"Error processing image: {str(e)}"
+        return f"**Error processing image:** {str(e)}"
 
-# Set up the Gradio interface.
+# Create the Gradio interface
 demo = gr.Interface(
     fn=solve_math_problem,
     inputs=gr.Image(
         type="pil", 
         label="Upload Handwritten Math Problem",
-        image_mode="RGB"  # This forces the input to be RGB.
+        image_mode="L"  # Grayscale mode improves OCR accuracy
     ),
-    outputs=gr.Textbox(label="Solution", show_copy_button=True),
+    outputs=gr.Markdown(),
     title="Handwritten Math Problem Solver",
-    description="Upload an image of a handwritten math problem (algebra, arithmetic, etc.) and get the solution",
+    description="Upload an image of a handwritten math problem, and the app will attempt to solve it.",
     examples=[
         ["example_addition.png"],
         ["example_algebra.jpg"]
     ],
-    theme="soft",
-    allow_flagging="never"
+    allow_flagging="never",
+    webpage_title="Handwritten Math Solver",
+    theme="soft"
 )
 
 if __name__ == "__main__":