Update app.py

app.py

@@ -0,0 +1,176 @@
+https://huggingface.co/spaces/hari6677/intrusion_detection/resolve/main/app.py
+import gradio as gr
+import tensorflow as tf
+import numpy as np
+import os
+
+print("🚀 Starting Intrusion Detection System...")
+
+# Load your trained model
+try:
+    model = tf.keras.models.load_model('improved_intrusion_detection_model.h5')
+    print("✅ Model loaded successfully!")
+except Exception as e:
+    print(f"❌ Model loading failed: {e}")
+    # Create a dummy model for testing
+    print("⚠️ Using dummy model for testing")
+
+def predict_intrusion(features_input):
+    """
+    Predict if network traffic is normal or attack
+    """
+    try:
+        # Convert input to array
+        features = [float(x.strip()) for x in features_input.split(',') if x.strip()]
+        
+        # Validate input length
+        if len(features) != 119:
+            return {
+                "prediction": "ERROR",
+                "confidence": 0,
+                "message": f"❌ Need exactly 119 features, but got {len(features)}"
+            }
+        
+        # Reshape for CNN model (1, 119, 1)
+        features_array = np.array(features).reshape(1, 119, 1)
+        
+        # Make prediction
+        prediction_prob = model.predict(features_array, verbose=0)[0][0]
+        confidence = float(prediction_prob)
+        
+        # Determine result
+        if confidence > 0.5:
+            result = {
+                "prediction": "🚨 ATTACK DETECTED",
+                "confidence": round(confidence * 100, 2),
+                "message": f"🚨 SECURITY ALERT! Potential intrusion detected with {confidence:.2%} confidence"
+            }
+        else:
+            result = {
+                "prediction": "✅ NORMAL TRAFFIC", 
+                "confidence": round((1 - confidence) * 100, 2),
+                "message": f"✅ Traffic appears normal with {(1-confidence):.2%} confidence"
+            }
+        
+        return result
+        
+    except Exception as e:
+        return {
+            "prediction": "ERROR",
+            "confidence": 0,
+            "message": f"❌ Error: {str(e)}"
+        }
+
+def generate_sample_features():
+    """Generate sample feature values for testing"""
+    # Normal traffic sample (mostly zeros and low values)
+    normal_features = [0.0] * 50 + [0.1, 0.2, 0.05, 0.0, 0.15] + [0.0] * 64
+    # Attack traffic sample (higher values)
+    attack_features = [0.8, 0.9, 0.7, 0.6, 0.85] + [0.0] * 50 + [0.9, 0.8, 0.95] + [0.0] * 61
+    
+    return {
+        "normal": ", ".join(map(str, normal_features)),
+        "attack": ", ".join(map(str, attack_features))
+    }
+
+# Generate samples
+samples = generate_sample_features()
+
+# Create Gradio interface
+with gr.Blocks(theme=gr.themes.Soft(), title="Network Intrusion Detection") as demo:
+    gr.Markdown(
+        """
+        # 🔒 Network Intrusion Detection System
+        **AI-Powered Threat Detection with 99.28% Accuracy**
+        """
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown("## 📊 Input Features")
+            
+            features_input = gr.Textbox(
+                label="Enter 119 Network Features (comma-separated)",
+                placeholder="0.0, 0.1, 0.2, 0.0, 0.15, ... (119 values total)",
+                lines=5
+            )
+            
+            with gr.Row():
+                analyze_btn = gr.Button("🚀 Analyze Traffic", variant="primary")
+                clear_btn = gr.Button("🗑️ Clear")
+            
+            gr.Markdown("### 🧪 Sample Data")
+            with gr.Row():
+                normal_btn = gr.Button("📋 Load Normal Sample")
+                attack_btn = gr.Button("⚠️ Load Attack Sample")
+        
+        with gr.Column():
+            gr.Markdown("## 📈 Prediction Results")
+            
+            prediction_output = gr.Label(
+                label="Detection Result",
+                value={"Prediction": "Waiting for input...", "Confidence": "0%"}
+            )
+            
+            message_output = gr.Textbox(
+                label="Security Alert",
+                interactive=False,
+                lines=3
+            )
+            
+            gr.Markdown("### 📊 Model Information")
+            gr.Markdown("""
+            - **Model Type**: CNN Deep Learning
+            - **Accuracy**: 99.28%
+            - **Attack Detection**: 99.40%
+            - **False Positive Rate**: 0.90%
+            - **Input Features**: 119
+            """)
+    
+    # Button actions
+    def load_normal_sample():
+        return samples["normal"]
+    
+    def load_attack_sample():
+        return samples["attack"]
+    
+    def clear_inputs():
+        return "", {"Prediction": "Waiting for input...", "Confidence": "0%"}, ""
+    
+    # Event handlers
+    analyze_btn.click(
+        fn=predict_intrusion,
+        inputs=features_input,
+        outputs=[prediction_output, message_output]
+    )
+    
+    normal_btn.click(
+        fn=load_normal_sample,
+        outputs=features_input
+    )
+    
+    attack_btn.click(
+        fn=load_attack_sample,
+        outputs=features_input
+    )
+    
+    clear_btn.click(
+        fn=clear_inputs,
+        outputs=[features_input, prediction_output, message_output]
+    )
+    
+    gr.Markdown(
+        """
+        ---
+        **How to use:**
+        1. Enter 119 comma-separated numerical values
+        2. Click "Analyze Traffic" to get prediction
+        3. Use sample buttons to test with pre-loaded data
+        
+        **Note**: This is a demo interface. For production use, ensure proper model deployment.
+        """
+    )
+
+# Launch the application
+if __name__ == "__main__":
+    demo.launch(debug=True)