Upload app.py with huggingface_hub

app.py

@@ -45,40 +45,18 @@ st.markdown('<p style="text-align: center; font-size: 1.2rem; color: #666;">Uplo
 
 @st.cache_resource
 def load_model():
-    """Load the trained model with compatibility fixes"""
+    """Load the Keras model - much more reliable!"""
     try:
-        model_path = 'Fast_VGG19_3epoch.h5'
+        model_path = 'Fast_VGG19_3epoch.keras'  # Changed to .keras format!
+        
         if os.path.exists(model_path):
-            # Try loading with custom objects to handle compatibility
-            try:
-                model = tf.keras.models.load_model(model_path, compile=False)
-                # Recompile the model
-                model.compile(
-                    optimizer='adam',
-                    loss='binary_crossentropy',
-                    metrics=['accuracy']
-                )
-                st.success("✅ Model loaded successfully!")
-                return model
-            except Exception as e1:
-                st.warning(f"First load attempt failed: {str(e1)}")
-                # Try alternative loading method
-                try:
-                    import h5py
-                    model = tf.keras.models.load_model(model_path, custom_objects=None, compile=False)
-                    model.compile(
-                        optimizer='adam',
-                        loss='binary_crossentropy',
-                        metrics=['accuracy']
-                    )
-                    st.success("✅ Model loaded with alternative method!")
-                    return model
-                except Exception as e2:
-                    st.error(f"Alternative load failed: {str(e2)}")
-                    return None
+            model = tf.keras.models.load_model(model_path)
+            st.success("✅ Keras model loaded successfully!")
+            return model
         else:
             st.error(f"Model file not found: {model_path}")
             return None
+            
     except Exception as e:
         st.error(f"Error loading model: {str(e)}")
         return None
@@ -86,19 +64,14 @@ def load_model():
 def preprocess_image(image):
     """Preprocess image for model prediction"""
     try:
-        # Convert to RGB array
         img = np.array(image.convert('RGB'))
-        
-        # Resize to model input size
         img = tf.image.resize(img, [64, 64])
         img = tf.cast(img, tf.float32) / 255.0
         
-        # Apply ImageNet normalization
         mean = [0.485, 0.456, 0.406]
         std = [0.229, 0.224, 0.225]
         img = (img - mean) / std
         
-        # Add batch dimension
         return tf.expand_dims(img, 0)
     
     except Exception as e:
@@ -112,9 +85,8 @@ with st.sidebar:
     **Architecture:** VGG19 Transfer Learning
     **Accuracy:** 77.0%
     **AUC Score:** 0.850
-    **Precision:** 75.0%
-    **Recall:** 22.2%
-    **Training Data:** RSNA Dataset
+    **Format:** Keras (.keras) ✅
+    **TensorFlow:** 2.15.0
     """)
     
     st.header("📋 How to Use")
@@ -124,19 +96,11 @@ with st.sidebar:
     3. **Click Analyze:** Get AI prediction
     4. **Review Results:** Check confidence levels
     """)
-    
-    st.header("🔧 Docker Info")
-    st.success("""
-    ✅ Running in Docker container
-    ✅ TensorFlow 2.15.0
-    ✅ Compatibility fixed
-    """)
 
 # Load model
 model = load_model()
 
 if model is not None:
-    # File uploader
     uploaded_file = st.file_uploader(
         "Choose a chest X-ray image",
         type=['jpg', 'jpeg', 'png'],
@@ -144,7 +108,6 @@ if model is not None:
     )
     
     if uploaded_file is not None:
-        # Create two columns for layout
         col1, col2 = st.columns([1, 1])
         
         with col1:
@@ -152,27 +115,21 @@ if model is not None:
             image = Image.open(uploaded_file)
             st.image(image, caption="Chest X-ray Image", use_column_width=True)
             
-            # Image info
             st.write(f"**Filename:** {uploaded_file.name}")
             st.write(f"**Size:** {image.size}")
-            st.write(f"**Format:** {image.format}")
         
         with col2:
             st.subheader("🤖 AI Analysis")
             
             if st.button("🔍 Analyze X-ray", type="primary", use_container_width=True):
                 with st.spinner("🧠 AI is analyzing the X-ray..."):
-                    # Preprocess image
                     processed_img = preprocess_image(image)
                     
                     if processed_img is not None:
                         try:
-                            # Make prediction
                             prediction = model.predict(processed_img, verbose=0)[0][0]
                             
-                            # Determine result
                             if prediction > 0.5:
-                                # Pneumonia detected
                                 confidence = prediction * 100
                                 pneumonia_html = f"""
                                 <div class="result-box pneumonia-result">
@@ -182,7 +139,6 @@ if model is not None:
                                 """
                                 st.markdown(pneumonia_html, unsafe_allow_html=True)
                                 
-                                # Risk level
                                 if confidence >= 80:
                                     st.error("🚨 **HIGH CONFIDENCE** - Immediate medical attention recommended")
                                 elif confidence >= 60:
@@ -191,7 +147,6 @@ if model is not None:
                                     st.info("ℹ️ **LOW CONFIDENCE** - Consider additional testing")
                             
                             else:
-                                # Normal result
                                 confidence = (1 - prediction) * 100
                                 normal_html = f"""
                                 <div class="result-box normal-result">
@@ -214,12 +169,10 @@ if model is not None:
                             normal_prob = (1 - prediction) * 100
                             pneumonia_prob = prediction * 100
                             
-                            # Progress bars
                             st.write("**Probability Breakdown:**")
                             st.progress(normal_prob/100, text=f"Normal: {normal_prob:.1f}%")
                             st.progress(pneumonia_prob/100, text=f"Pneumonia: {pneumonia_prob:.1f}%")
                             
-                            # Numerical display
                             col_a, col_b = st.columns(2)
                             with col_a:
                                 st.metric("Normal", f"{normal_prob:.1f}%")
@@ -236,7 +189,6 @@ if model is not None:
 
 else:
     st.error("❌ Failed to load the AI model. Please check the deployment.")
-    st.info("💡 Try refreshing the page or contact support if the issue persists.")
 
 # Footer
 footer_html = """
@@ -249,7 +201,7 @@ footer_html = """
     <hr style="margin: 1rem 0;">
     
     <p><em>🐳 Powered by Docker • TensorFlow 2.15.0 • Streamlit</em></p>
-    <p><em>🤖 VGG19 Transfer Learning Model</em></p>
+    <p><em>🤖 VGG19 Transfer Learning Model (.keras format)</em></p>
 </div>
 """