Table 3 MUSDTI hyperparameter values.

From: Reciprocal perspective as a super learner improves drug-target interaction prediction (MUSDTI)

MUSDTI model parameter

DeepDTA dataset

DeepDTA* dataset

Double-cold dataset

Double-cold* dataset

Colsample by tree

0.9362

0.7991

0.8988

0.9278

Gamma

1.1306

2.918

1.9723

4.342

Learning rate

0.2637

0.095

1.748

0.093

Max depth

15.0

13.0

12.0

12.0

Min child weight

2.0

9.000

4.0

1.0

  1. The * designation denotes the model parameters used for the ensembled MUSDTI* model prior to the cascaded application of RP.
Back to article page