Weird test accuracy curve

[howiethepeanut]

I tested your original codes and found out that the test accuracy dropped sharply around 40 epochs and fluctuated drastically until the final 20 epochs which does not match the curve in your paper. Can you explain why this happened?

[bbdamodaran]

I also obtain the similar behavior with cifar10 dataset, I couldn't reproduce results as shown in the paper.
I didn't change anything from the code, as the epoch increases the accuracy drops and again increases after 170 epochs, and the accuracy curve is not stable, as observed by @howiethepeanut

[bbdamodaran]

@howiethepeanut @bhanML If you have the pytorch implementation of MentorNet, can you please share with me
thanks

[devraj89]

Hi

I am also getting the same problem. The Co-teaching learning accuracies are not so consistent and wildly varies over the epochs and starts becoming consistent around epoch ~180.
@howiethepeanut @bbdamodaran Did You find any particular reason for this ?

Thanks
Devraj