Table 2 Consensus when removing models trained on the associated test dataset

1

1050

0.663

0.952

0.706

1751

0.205

0.968

0.867

2

857

0.712

0.835

0.612

1472

0.218

0.865

0.865

3

578

0.811

0.642

0.557

947

0.244

0.623

0.843

4

357

0.868

0.424

0.482

490

0.282

0.372

0.828

5

180

0.911

0.224

0.427

136

0.331

0.121

0.809

6

81

0.926

0.103

0.397

29

0.276

0.022

0.800

7

30

0.967

0.040

0.384

3

0.333

0.003

0.799

8

9

0.889

0.011

0.377

-

-

-

-

9

1

1.000

0.001

0.375

-

-

-

-