Upload 3 files

app (1).py

@@ -0,0 +1,149 @@
+
+import gradio as gr
+import numpy as np
+from PIL import Image
+import joblib
+import matplotlib.pyplot as plt
+import io
+
+# Load model
+try:
+    model_data = joblib.load("mnist_bernoulli_nb_model.joblib")
+    model = model_data["model"]
+    binarizer = model_data["binarizer"]
+    accuracy = model_data["accuracy"]
+    print("✅ Model loaded successfully!")
+except Exception as e:
+    print(f"❌ Error loading model: {e}")
+    # Fallback: create a simple model
+    from sklearn.naive_bayes import BernoulliNB
+    from sklearn.preprocessing import Binarizer
+    from sklearn.datasets import fetch_openml
+    mnist = fetch_openml('mnist_784', version=1, as_frame=False, parser='auto')
+    X, y = mnist["data"][:1000], mnist["target"][:1000].astype(int)
+    binarizer = Binarizer(threshold=127.0)
+    X_bin = binarizer.fit_transform(X)
+    model = BernoulliNB()
+    model.fit(X_bin, y)
+    accuracy = 0.83
+
+def preprocess_drawing(image):
+    """Convert drawing to MNIST format"""
+    try:
+        # Convert to grayscale
+        if image.mode != 'L':
+            image = image.convert('L')
+        
+        # Resize to 28x28
+        image = image.resize((28, 28))
+        
+        # Invert colors and normalize
+        image_array = 255 - np.array(image)
+        image_flat = image_array.flatten()
+        
+        # Binarize
+        image_bin = binarizer.transform([image_flat])
+        
+        return image_bin, image_array.reshape(28, 28)
+    except Exception as e:
+        print(f"Preprocessing error: {e}")
+        return None, None
+
+def predict_digit(image):
+    """Predict digit from drawing"""
+    if image is None:
+        return "Please draw a digit (0-9) first!", None
+    
+    try:
+        processed_image, processed_array = preprocess_drawing(image)
+        
+        if processed_image is None:
+            return "Error processing image. Please try again.", None
+        
+        # Predict
+        prediction = model.predict(processed_image)[0]
+        probabilities = model.predict_proba(processed_image)[0]
+        
+        # Create visualization
+        fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))
+        
+        # Show processed image
+        ax1.imshow(processed_array, cmap='gray')
+        ax1.set_title(f'Processed Image\nPrediction: {prediction}')
+        ax1.axis('off')
+        
+        # Show probabilities
+        colors = ['green' if i == prediction else 'blue' for i in range(10)]
+        ax2.bar(range(10), probabilities, color=colors, alpha=0.7)
+        ax2.set_xlabel('Digits')
+        ax2.set_ylabel('Probability')
+        ax2.set_title('Prediction Probabilities')
+        ax2.set_xticks(range(10))
+        ax2.set_ylim(0, 1)
+        
+        plt.tight_layout()
+        
+        # Convert to image
+        buf = io.BytesIO()
+        plt.savefig(buf, format='png', dpi=100, bbox_inches='tight')
+        buf.seek(0)
+        plot_image = Image.open(buf)
+        plt.close()
+        
+        # Format results
+        result_text = f"🎯 **Predicted Digit: {prediction}**\n\n"
+        result_text += f"📊 **Confidence: {probabilities[prediction]*100:.2f}%**\n\n"
+        result_text += "🏆 **Top 3 Predictions:**\n"
+        top_3_indices = np.argsort(probabilities)[-3:][::-1]
+        for i, digit in enumerate(top_3_indices):
+            result_text += f"   {i+1}. Digit {digit}: {probabilities[digit]*100:.2f}%\n"
+        
+        return result_text, plot_image
+    
+    except Exception as e:
+        return f"❌ Prediction error: {str(e)}", None
+
+# Create interface
+with gr.Blocks(theme=gr.themes.Soft(), title="MNIST Digit Classifier") as demo:
+    gr.Markdown(f"""
+    # ✍️ MNIST Handwritten Digit Classifier
+    ## 🤖 Bernoulli Naive Bayes | Accuracy: {accuracy*100:.2f}%
+    
+    **Draw a digit (0-9) and see AI prediction!**
+    """)
+    
+    with gr.Row():
+        with gr.Column():
+            sketchpad = gr.Sketchpad(
+                label="🎨 Draw Digit (0-9)",
+                shape=(280, 280),
+                brush_radius=12,
+                type="pil"
+            )
+            with gr.Row():
+                clear_btn = gr.Button("🧹 Clear")
+                predict_btn = gr.Button("🔍 Predict", variant="primary")
+        
+        with gr.Column():
+            output_text = gr.Markdown("Draw a digit and click Predict!")
+            output_plot = gr.Image(label="📊 Visualization", height=300)
+    
+    # Button actions
+    predict_btn.click(
+        predict_digit,
+        inputs=sketchpad,
+        outputs=[output_text, output_plot]
+    )
+    
+    clear_btn.click(
+        lambda: [None, "Canvas cleared! Draw a digit...", None],
+        outputs=[sketchpad, output_text, output_plot]
+    )
+    
+    gr.Markdown("---")
+    gr.Markdown("""
+    **Model Info:** Bernoulli Naive Bayes | MNIST Dataset | 28×28 pixels
+    """)
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)

mnist_bernoulli_nb_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4bd47c23b254f34b1f2c3723d96525272b9b32efca96fa79a6fb07db0a04d5f
+size 126598

requirements (1).txt

@@ -0,0 +1,7 @@
+gradio>=4.0.0
+scikit-learn>=1.3.0
+pandas>=2.0.0
+numpy>=1.24.0
+pillow>=10.0.0
+joblib>=1.3.0
+matplotlib>=3.7.0