Evaluating Performances - Regression Error Metrics

Regression is a task when a model attempts to predict continuous values (for categorical values - classification). most common evaluation metrics for regression; Mean Absolute Error (MAE) : the mean of the absulute values of errors. won't punish large errors. Mean Squared Error (MSE) : larger errors are noted more than MAE, making MSE more popular. but it squares the units itself not only the differences, so we use RMSE Root Mean Square Error (RMSE) : root of MSE. the most popular....