Skip to main content

Table 1 Number of data instances used for training and validation after removal of all-zero value rows

From: Empowering the discovery of novel target-disease associations via machine learning approaches in the open targets platform

Set

Fold1

Fold2

Fold3

Fold4

Fold5

Train

     

Positive

15,137

14,382

15,120

14,435

14,918

Negative

70,945

67,020

70,210

73,575

71,941

Total

86,082

81,402

85,330

88,010

86,859

Validation

     

Positive

3369

4085

3404

4098

3561

Negative

18,132

20,313

18,424

15,344

16,194

Total

21,501

24,398

21,828

19,442

19,755

  1. Held-out testing data comprised of 46,290 instances (7382 positive: 38,907 negative)
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us