Table 2 Optimized values of the XGB hyper-parameters.
From: Sea-surface pCO2 maps for the Bay of Bengal based on advanced machine learning algorithms
Hyper-parameters | Range or Options | Optimized Value |
|---|---|---|
lambda | 0–1.0 | 0.8634 |
alpha | 0-1 | 0.2574 |
subsample | 0-1 | 0.6920 |
Booster | gbtree/gblinear/dart | gbtree |
colsample_bytree | 0-1 | 0.6460 |
max_depth | 10–100 (step = 1) | 93 |
min_child_weight | 1–100 | 36 |
learning_rate | 1 Ă— e-08-1 | 0.0001 |
gamma | 1 Ă— e-08-1 | 5.546 Ă— e-07 |
n_estimators | 100–150 (step = 1) | 131 |
grow_policy | depthwise/lossguide | lossguide |
