Table 3 Optimal hyper parameter used for models in both category.

From: Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India

Models

Hyper-parameters

Category-I

Category-II

Search range

Optimal value

Search range

Optimal value

XGBoost

nrounds

333, 633, 933

633

333, 633, 933

933

Max depths

6, 15, 18

15

6, 15, 18

15

η

0.3, 0.5, 0.8, 1

0.5

0.3, 0.5, 0.8, 1

0.5

γ

0, 0.5, 1

0

0, 0.5, 1

0

Sub sample

0.5, 0.75, 1

1

0.5, 0.75, 1

1

RF

Mtry

15

13

15

13

Ntree

500, 933, 1333

933

500, 933, 1333

933

KNN

K- neighbors

0–50 with the increment of 1

5

0–50 with the increment of 1

5