Update app.py

app.py

@@ -1,46 +1,58 @@
 import pandas as pd
 import joblib
 import gradio as gr
+import seaborn as sns
+import matplotlib.pyplot as plt
 
 # Load the trained model
 model = joblib.load("anomaly_detector_rf_model.pkl")
 
-# Define feature columns (include all used during training)
-feature_cols = ['hour', 'day_of_week', 'is_weekend', 'amount', 'merchant_avg_amount',
-                'amount_zscore', 'log_amount', 'type_atm_withdrawal', 'type_credit',
-                'type_debit', 'merchant_encoded']
+# Define feature order
+feature_order = ['hour', 'day_of_week', 'is_weekend', 'amount', 'merchant_avg_amount', 
+                 'amount_zscore', 'log_amount', 'type_atm_withdrawal', 'type_credit', 
+                 'type_debit', 'merchant_encoded']
 
-def detect_anomalies(file_path):
-    # Read the dataset
-    df = pd.read_csv(file_path)
+def detect_anomalies(data):
+    df = pd.DataFrame(data)
+    df = df[feature_order]  # Ensure correct feature order
+    df['is_anomalous'] = model.predict(df)
     
-    # Ensure all features exist in the dataframe
-    missing_cols = [col for col in feature_cols if col not in df.columns]
-    if missing_cols:
-        return f"Missing columns in dataset: {missing_cols}"
+    # Filter anomalies and display relevant details
+    anomalies = df[df['is_anomalous'] == 1][['transaction_id', 'merchant', 'location', 'amount']]
+    return anomalies
+
+# Function to generate plots
+def generate_plots(df):
+    fig, axes = plt.subplots(2, 2, figsize=(12, 10))
     
-    # Align feature order with model training
-    df = df[feature_cols]
+    sns.countplot(data=df, x='is_anomalous', palette='Set2', ax=axes[0, 0])
+    axes[0, 0].set_title("Anomaly Distribution")
     
-    # Make predictions
-    df['is_anomalous'] = model.predict(df)
+    sns.countplot(data=df, y='merchant', order=df['merchant'].value_counts().index, palette='viridis', ax=axes[0, 1])
+    axes[0, 1].set_title("Transactions by Merchant")
     
-    # Filter anomalous transactions
-    anomalies = df[df['is_anomalous'] == 1][['transaction_id', 'merchant', 'location', 'amount']]
+    sns.histplot(df['amount'], bins=30, kde=True, color='blue', ax=axes[1, 0])
+    axes[1, 0].set_title("Transaction Amount Distribution")
     
-    # Save to a new CSV
-    anomalies.to_csv("predicted_anomalies.csv", index=False)
+    sns.scatterplot(data=df, x='amount', y='merchant_avg_amount', hue='is_anomalous', palette='coolwarm', ax=axes[1, 1])
+    axes[1, 1].set_title("Amount vs. Merchant Average Amount")
     
-    return anomalies
+    plt.tight_layout()
+    return fig
 
 # Gradio Interface
+def app_interface(file):
+    df = pd.read_csv(file.name)
+    anomalies = detect_anomalies(df)
+    plot = generate_plots(df)
+    return anomalies, plot
+
 interface = gr.Interface(
-    fn=detect_anomalies,
-    inputs=gr.File(label="Upload CSV File"),
-    outputs=gr.Dataframe(label="Predicted Anomalies"),
-    title="Anomaly Detection System",
-    description="Upload a transaction dataset to detect anomalies."
+    fn=app_interface,
+    inputs=[gr.File(label="Upload Transaction Data (CSV)")],
+    outputs=[gr.Dataframe(label="Detected Anomalies"), gr.Plot(label="Transaction Analysis Charts")],
+    title="Financial Anomaly Detection",
+    description="Upload a transaction dataset to detect financial anomalies and visualize transaction patterns."
 )
 
-if __name__ == "__main__":
-    interface.launch(share=True)
+interface.launch(share=True)