Update app.py

app.py

@@ -15,24 +15,24 @@ MODEL_NAMES = [
     'XGBoost Regressor',
     'LGBM Regressor',
 ]
-
-def load_models(model_names: List[str]) -> Dict[str, Any]:
-    """Load machine learning models from disk."""
-    models = {}
-    for name in model_names:
-        path = os.path.join(MODEL_DIR, f"{name.replace(' ', '')}.joblib")
-        try:
-            models[name] = joblib.load(path)
-        except Exception as e:
-            st.error(f"Error loading model {name}: {str(e)}")
-    return models
+  
+@st.cache
+def load_data(path: str) -> pd.DataFrame:
+    """Load the dataset with caching."""
+    return pd.read_csv(path)
 
 # Load models
 models = load_models(MODEL_NAMES)
 
 # Load dataset
 data_path = os.path.join(DATA_DIR, DATA_FILE)
-df = pd.read_csv(data_path)
+
+# Load dataset with error handling
+try:
+    df = load_data(data_path)
+except Exception as e:
+    st.error(f"Error loading dataset: {str(e)}")
+    st.stop()
 
 # Prepare features and target
 X = df.drop(columns=['Salary'])
@@ -56,6 +56,7 @@ max_years = float(df.YearsOfExperience.max() * 1.5)
 # Precompute predictions for training set
 y_train_predictions = {name: model.predict(X_train) for name, model in models.items()}
 
+# Include more metrics for model performance
 def load_and_predict(sample: pd.DataFrame) -> pd.DataFrame:
     """Predict salary using loaded models and evaluate statistics."""
     results = []
@@ -63,10 +64,14 @@ def load_and_predict(sample: pd.DataFrame) -> pd.DataFrame:
     for name, model in models.items():
         try:
             salary_pred = model.predict(sample)[0]
+            mae = mean_absolute_error(y_train, y_train_predictions[name])  # Example metric
+            mse = mean_squared_error(y_train, y_train_predictions[name])  # Example metric
             results.append({
                 'Model': name,
                 'Predicted Salary': salary_pred,
                 'R2 Score (%)': r2_score(y_train, y_train_predictions[name]) * 100,
+                'Mean Absolute Error': mae,
+                'Mean Squared Error': mse,
             })
         except Exception as e:
             st.error(f"Error during prediction with model {name}: {str(e)}")