Aboute learning MentorNet DD on small clean CIFAR-10 subset and apply it to CIFAR-100

[ruirui88]

In the supplementary of paper, it writes that as CIFAR-100 and CIFAR-10 have the different number of classes, to apply a MentorNet, we fix the class label to 0. It's not clear which label is fix to 0, because there are two labes for samples, i.e., clean labels and noisy labels.

[roadjiang]

That refers to the input label to 0 to the mentornet. See: https://github.com/google/mentornet/blob/76d6be2db1be39714dec6db6bb3bcbb77855ce6e/code/cifar_train_mentornet.py#L205

On Tue, Sep 10, 2019 at 1:49 AM ruirui88 notifications@github.com wrote:

In the supplementary of paper, it writes that as CIFAR-100 and CIFAR-10 have the different number of classes, to apply a MentorNet, we fix the class label to 0. It's not clear which label is fix to 0, because there are two labes for samples, i.e., clean labels and noisy labels.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/google/mentornet/issues/9?email_source=notifications&email_token=ADGNQMMT4NGKHPRLSBC4MRDQI5NRNA5CNFSM4IVEZEUKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HKMH6RQ, or mute the thread https://github.com/notifications/unsubscribe-auth/ADGNQMIEPR77TW2PWIKR5B3QI5NRNANCNFSM4IVEZEUA .

-- Have a good day Lu Jiang

[ruirui88]

I got it, thanks.