Update app.py

app.py

@@ -12,6 +12,15 @@ from sklearn.metrics import precision_score, recall_score
 from datetime import datetime
 
 def generate_report(data, predictions, model_names, metrics):
+    # Create markdown table manually
+    metrics_table = "\n".join([
+        "| Model | Precision | Recall |",
+        "|-------|-----------|--------|"
+    ] + [
+        f"| {row['Model']} | {row['Precision']} | {row['Recall']} |" 
+        for _, row in metrics.iterrows()
+    ])
+    
     report = f"""
     Network Anomaly Detection Report
     Generated: {datetime.now().strftime("%Y-%m-%d %H:%M:%S")}
@@ -24,7 +33,7 @@ def generate_report(data, predictions, model_names, metrics):
     - Anomaly Percentage: {sum(predictions)/len(data):.2%}
     
     Model Performance:
-    {metrics.to_markdown()}
+    {metrics_table}
     
     Conclusion:
     The system detected {sum(predictions)} potential anomalies using ensemble of {len(model_names)} models.
@@ -65,12 +74,16 @@ def main():
     contamination = st.sidebar.slider("Expected Anomaly Ratio", 0.01, 0.5, 0.1)
     ensemble_method = st.sidebar.selectbox("Ensemble Method", ["Average", "MOA", "AOM"])
 
-    # Data input section
+    # Data handling
     uploaded_file = st.file_uploader("Upload network data (CSV)", type=["csv"])
     
     if uploaded_file:
-        data = pd.read_csv(uploaded_file)
-        st.success("Uploaded data loaded successfully!")
+        try:
+            data = pd.read_csv(uploaded_file)
+            st.success("Uploaded data loaded successfully!")
+        except Exception as e:
+            st.error(f"Error reading file: {str(e)}")
+            return
     else:
         # Generate synthetic network data
         np.random.seed(42)
@@ -90,6 +103,10 @@ def main():
 
     # Data preprocessing
     numeric_cols = data.select_dtypes(include=np.number).columns.tolist()
+    if not numeric_cols:
+        st.error("No numeric columns found for analysis!")
+        return
+        
     X = data[numeric_cols].values
     X_norm = standardizer(X)
 
@@ -100,20 +117,24 @@ def main():
         "One-Class SVM": OCSVM(contamination=contamination)
     }
 
-    selected_models = [model_dict[m] for m in models]
+    selected_models = [model_dict[m] for m in models if m in model_dict]
     if not selected_models:
         st.error("Please select at least one detection model!")
         return
 
-    # Training and prediction
-    st.subheader("Model Training Progress")
-    progress_bar = st.progress(0)
-    train_scores = np.zeros([len(X), len(selected_models)])
-
-    for i, model in enumerate(selected_models):
-        model.fit(X_norm)
-        train_scores[:, i] = model.decision_function(X_norm)
-        progress_bar.progress((i+1)/len(selected_models))
+    # Model training
+    try:
+        st.subheader("Model Training Progress")
+        progress_bar = st.progress(0)
+        train_scores = np.zeros([len(X), len(selected_models)])
+
+        for i, model in enumerate(selected_models):
+            model.fit(X_norm)
+            train_scores[:, i] = model.decision_function(X_norm)
+            progress_bar.progress((i+1)/len(selected_models))
+    except Exception as e:
+        st.error(f"Model training failed: {str(e)}")
+        return
 
     # Ensemble prediction
     try:
@@ -157,8 +178,8 @@ def main():
 
     # Visualization
     st.subheader("Data Visualization")
-    
     tab1, tab2 = st.tabs(["2D Projection", "3D Projection"])
+    
     with tab1:
         fig, ax = plt.subplots(figsize=(10, 6))
         pca = PCA(n_components=2)
@@ -176,12 +197,12 @@ def main():
     with tab2:
         st.pyplot(plot_3d_projections(X_norm, predictions))
 
-    # Generate report
+    # Report generation
     st.subheader("Analysis Report")
     report = generate_report(data[numeric_cols], predictions, models, metrics_df)
     st.code(report, language='text')
 
-    # Report download
+    # Download report
     st.download_button(
         label="Download Full Report",
         data=report,