Questions about the "10-fold cross validation".

[YuanweiLee1]

Hello! Thank you for your great work. When I try to replicate the experiment on MoNuSeg dataset, a problem occurs in analyzing the code of 10-fold cross validation e.g I can't find where you implement the 10-fold cross validation. Because I find that you split the dataset into training, validation, and test sets in "_datasplit" function of "my_dataset.py", but this is done only once instead of 10 times. Also in "train.py", the model is trained only once instead of 10 times with different dataset split at the require of 10-fold cross validation. Could you please tell me how you implement 10-fold cross validation and release the code? Best regards.

[RuoyuGuo]

Hi, if you look at the 207-208 line intrain.py, there is a parameter namedk_fold. You can use a different value ofk_fold to run train.py 10 times to get 10-fold cross validation results.

[YuanweiLee1]

Hi, if you look at the 207-208 line intrain.py, there is a parameter namedk_fold. You can use a different value ofk_fold to run train.py 10 times to get 10-fold cross validation results.

I get it, thanks a lot!