Table 3 Model evaluation metrics for Multi-parameter coupling Compensation.
Metric name | Calculation formula | Evaluation standard | Weight coefficient |
|---|---|---|---|
Mean Absolute Error | \(\:MAE=\frac{1}{n}\sum\:_{i=1}^{n}\parallel\:{y}_{i}-{\widehat{y}}_{i}\parallel\:\) | < 0.5% excellent | 0.30 |
Root Mean Square Error | \(\:RMSE=\sqrt{\frac{1}{n}\sum\:_{i=1}^{n}{\left({y}_{i}-{\widehat{y}}_{i}\right)}^{2}}\) | < 0.3% excellent | 0.25 |
Maximum Absolute Error | \(\:MA{E}_{max}={\text{m}\text{a}\text{x}}_{i}\parallel\:{y}_{i}-{\widehat{y}}_{i}\parallel\:\) | < 2.0% acceptable | 0.20 |
Correlation Coefficient | \(\:R=\frac{\sum\:\left({y}_{i}-\stackrel{-}{y}\right)\left({\widehat{y}}_{i}-\stackrel{-}{\widehat{y}}\right)}{\sqrt{\sum\:{\left({y}_{i}-\stackrel{-}{y}\right)}^{2}\sum\:{\left({\widehat{y}}_{i}-\stackrel{-}{\widehat{y}}\right)}^{2}}}\) | > 0.95 excellent | 0.15 |
Computational Time | \(\:{T}_{comp}=\frac{{t}_{inference}}{{t}_{baseline}}\) | < 1.5 acceptable | 0.10 |
