Table 4 Computational time comparison between the models.
From: Parallel boosting neural network with mutual information for day-ahead solar irradiance forecasting
Location | Models | Convergence time | Inference time | Computational time |
|---|---|---|---|---|
(s) | (s) | (s) | ||
Islamabad | CatBoost | 0.459 | 0.033 | 0.492 |
XgBoost | 0.247 | 0.003 | 0.25 | |
RF | 0.285 | 0.009 | 0.294 | |
LSTM | 284.637 | 0.466 | 285.103 | |
GRU | 243.045 | 0.471 | 243.516 | |
PBNN | 9.906 | 0.084 | 9.99 | |
San Diego | CatBoost | 0.308 | 0.001 | 0.309 |
XgBoost | 0.381 | 0.002 | 0.383 | |
RF | 0.144 | 0.002 | 0.146 | |
LSTM | 56.238 | 0.292 | 56.53 | |
GRU | 50.159 | 0.229 | 50.388 | |
PBNN | 3.342 | 0.036 | 3.379 |
