feat: updated app.py

app.py

@@ -274,37 +274,41 @@ elif page == "Customer Analysis":
                 st.write("Feature names:")
                 st.write(gbm.feature_name())
 
-                # Load X_predict for that cluster
-                X_predict_cluster = pd.read_csv(f'predicts/X_predict_cluster_{cluster}.csv')
+                # Load predict data for that cluster
+                predict_data = pd.read_csv(f'predicts/predict_cluster_{cluster}.csv')
                 
                 # Convert cliente_id to string
-                X_predict_cluster['cliente_id'] = X_predict_cluster['cliente_id'].astype(str)
+                predict_data['cliente_id'] = predict_data['cliente_id'].astype(str)
                 
-                st.write("### X_predict_cluster DataFrame:")
-                st.write(X_predict_cluster.head())
-                st.write(f"Shape: {X_predict_cluster.shape}")
+                st.write("### Predict Data DataFrame:")
+                st.write(predict_data.head())
+                st.write(f"Shape: {predict_data.shape}")
 
                 # Filter for the specific customer
                 customer_code_str = str(customer_code)
-                X_cliente = X_predict_cluster[X_predict_cluster['cliente_id'] == customer_code_str]
+                customer_data = predict_data[predict_data['cliente_id'] == customer_code_str]
                 
                 # Add debug statements
-                st.write(f"Unique customer IDs in X_predict_cluster: {X_predict_cluster['cliente_id'].unique()}")
+                st.write(f"Unique customer IDs in predict data: {predict_data['cliente_id'].unique()}")
                 st.write(f"Customer code we're looking for: {customer_code_str}")
 
-                st.write("### X_cliente DataFrame:")
-                st.write(X_cliente.head())
-                st.write(f"Shape: {X_cliente.shape}")
+                st.write("### Customer Data:")
+                st.write(customer_data.head())
+                st.write(f"Shape: {customer_data.shape}")
 
-                if not X_cliente.empty:
+                if not customer_data.empty:
+                    # Define features consistently with the training process
+                    lag_features = [f'precio_total_lag_{lag}' for lag in range(1, 25)]
+                    features = lag_features + ['mes', 'marca_id_encoded', 'año', 'cluster_id']
+                    
                     # Prepare data for prediction
-                    features_for_prediction = X_cliente.drop(columns=['cliente_id', 'fecha_mes'])
+                    X_predict = customer_data[features]
                     st.write("### Features for Prediction:")
-                    st.write(features_for_prediction.head())
-                    st.write(f"Shape: {features_for_prediction.shape}")
+                    st.write(X_predict.head())
+                    st.write(f"Shape: {X_predict.shape}")
                     
                     # Make Prediction for the selected customer
-                    y_pred = gbm.predict(features_for_prediction, num_iteration=gbm.best_iteration)
+                    y_pred = gbm.predict(X_predict, num_iteration=gbm.best_iteration)
                     st.write("### Prediction Results:")
                     st.write(f"Type of y_pred: {type(y_pred)}")
                     st.write(f"Shape of y_pred: {y_pred.shape}")
@@ -312,7 +316,7 @@ elif page == "Customer Analysis":
                     st.write(y_pred[:5])
                     
                     # Reassemble the results
-                    results = X_cliente[['cliente_id', 'marca_id_encoded', 'fecha_mes']].copy()
+                    results = customer_data[['cliente_id', 'marca_id_encoded', 'fecha_mes']].copy()
                     results['ventas_predichas'] = y_pred
                     st.write("### Results DataFrame:")
                     st.write(results.head())