From: Shapley-based interpretation of deep learning models for wildfire spread rate prediction
Metric | Interpretation |
---|---|
Mean absolute error (MAE) | - This metric quantifies the average absolute deviations between expected and actual values. - The metric under consideration exhibits a higher degree of robustness in the presence of outliers as compared to the root mean square error (RMSE). |
Root mean square error (RMSE) | - Emphasizes squared difference between predicted and actual values. - Gives greater weight to outliers compared to MAE. |
Mean bias error (MBE) | - Quantifies average bias in model’s predictions. - Indicates systematic overestimation (MBE > 0) or underestimation (MBE < 0). - Takes into account the direction of error. - Value close to zero denotes minimal bias. |