Table 3 The optimal parameters of XGBoost.

From: Visible-NIR hyperspectral classification of grass based on multivariate smooth mapping and extreme active learning approach

Parameter

Setting

Booster

gbtree

N estimators

160

Max depth

5

Min child weight

1

Subsample

0.6

Colsample bytree

0.6

Reg alpha

1e−05

Reg lambda

1

Eta

0.1

Learning rate

0.1

Nthread

4

Scale pos weight

1

Seed

27

Num. class

15

Back to article page