The Root Mean Squared Error is equivalent to the standard deviation of the error residuals in a regression problem. Since RMSE is just the square root of the MSE, RMSE is also sensitive to outliers because larger errors have a disproportionately large effect on the score.
Note: In order to maintain the convention of maximizing validation scores, this metric outputs the negative of the original score.
Estimator Compatibility: Regressor
Output Range: -∞ to 0
use Rubix\ML\CrossValidation\Metrics\RMSE; $metric = new RMSE();