Regression Metrics
From now on:
\(\hat{y}\) is the estimated target output,
\(y\) is the corresponding (correct) target output.
\(\hat{Y}\) is the whole estimated target output ,
\(Y\) is the corresponding (correct) target output.
\(mean(\hat{Y})\) is the mean of whole estimated target output ,
\(mean(Y)\) is the mean of whole (correct) target output.
- EVS - Explained Variance Score
- ME - Max Error
- MAE - Mean Absolute Error
- MSE - Mean Squared Error
- RMSE - Root Mean Square Error
- MSLE - Mean Squared Logarithmic Error
- MedAE - Median Absolute Error
- MRE - Mean Relative Error
- MAPE - Mean Absolute Percentage Error
- SMAPE - Symmetric Mean Absolute Percentage Error
- MAAPE - Mean Arctangent Absolute Percentage Error
- MASE - Mean Absolute Scaled Error
- NSE - Nash-Sutcliffe Efficiency
- WI - Willmott Index
- R - Pearson’s Correlation Index
- R2 - Coefficient of Determination
- CI - Confidence Index
- R2s - (Pearson’s Correlation Index)**2
- DRV - Deviation of Runoff Volume
- KGE - Kling-Gupta Efficiency
- GINI - GINI Coefficient
- PCD - Prediction of Change in Direction
- E - Entropy Loss
- CE - Cross Entropy
- KLD - Kullback-Leibler Divergence
- JSD - Jensen-Shannon Divergence
- VAF - Variance Accounted For
- RAE - Relative Absolute Error
- A10 - A10 index
- A20 - A20 index
- NRMSE - Normalized Root Mean Square Error
- RSE - Residual Standard Error
- RE - Relative Error
- AE - Absolute Error
- SE - Squared Error
- SLE - Squared Log Error