search

ChengF-Lab / deepDR

50 stars 31 forks source link

Lack of remove edges after cross-fold validation #6

Closed dlutwy closed 3 years ago

dlutwy commented 3 years ago

According to the available code, the positive instances of eval & test data will be seen in training data. They should be removed after splitting for accurate performance metrics during cross-validation.

dlutwy commented 3 years ago

sorry, I misunderstood the code.