Sklearn Random Forest Regression Example
Python
An example of training a Random Forest model using RandomForestRegressor from Sklearn, evaluating the model on test data and reviewing feature importance.
1| from sklearn.ensemble import RandomForestRegressor 2| from sklearn.metrics import mean_squared_error, mean_absolute_error, max_error, explained_variance_score, mean_absolute_percentage_error 3| 4| # Step 1: Initialise & fit Random Forest Regressor 5| model = RandomForestRegressor(n_estimators=10, 6| criterion="squared_error", 7| max_depth=None, 8| min_samples_split=2, 9| min_samples_leaf=1, 10| n_jobs=-1, 11| random_state=101) 12| model.fit(X_train, y_train) 13| 14| # Step 2: Create dictionary that contains feature importance 15| feature_importance= dict(zip(X_train.columns, model.feature_importances_)) 16| print('Feature Importance',feature_importance) 17| 18| # Step 3: Make prediction for test data & evaluate performance 19| y_pred = model.predict(X_test) 20| print('RMSE:',mean_squared_error(y_test, y_pred, squared = False)) 21| print('MAE:',mean_absolute_error(y_test, y_pred)) 22| print('MAPE:',mean_absolute_percentage_error(y_test, y_pred)) 23| print('Max Error:',max_error(y_test, y_pred)) 24| print('Explained Variance Score:',explained_variance_score(y_test, y_pred))
133
121
117
110