Skip to main content

Table 2 Gradient boosting hyper-parameters chosen by grid search for each model

From: Prediction of plant lncRNA by ensemble machine learning classifiers

GB Model #

Learning rate

Maxdepth

Subsample

n estimators

1

0.04

10

0.6

100

2

0.04

10

0.6

100

3

0.04

10

0.6

100

4

0.02

8

0.6

100

5

0.02

10

0.6

100

6

0.02

10

0.6

100

7

0.04

10

0.6

100

8

0.04

10

0.6

100

  1. Hyper-parameters were chosen by grid search using 30 iterations of 4-fold nested cross validation. The given hyper-parameters corresponded to models with the highest accuracy values of all given hyper-parameter combinations
Back to article page

BMC Genomics

ISSN: 1471-2164

Contact us