problem about N fold cross validation using create_split_seq.py

mahmoodlab / CLAM

Data-efficient and weakly supervised computational pathology on whole slide images - Nature Biomedical Engineering

http://clam.mahmoodlab.org

GNU General Public License v3.0

1.02k stars 340 forks source link

problem about N fold cross validation using create_split_seq.py #197

Closed ljhOfGithub closed 11 months ago

ljhOfGithub commented 1 year ago

If I use the command in the graph,I will get ten split files,but there are not enough data in every file,because you set the val_frac and test_frac as 0.1,which will make the portion of the remaining samples are (1-0.1-0.1)*0.75=0.6,losing 20% original data.

ShiCrazy commented 1 year ago

You can just assign a value "1" to "label_frac"