search

vikasverma1077 / manifold_mixup

Code for reproducing Manifold Mixup results (ICML 2019)
483 stars 65 forks source link

Question about accuracy during Training #9

Closed saikalyan9981 closed 4 years ago

saikalyan9981 commented 5 years ago

I was trying to implement mix-up-hidden for preactresnet18, I found that for calculating accuracy you are comparing mix-up output with original target instead of re-weighted one, that makes accuracy low during training, I didn't understand what that accuracy signifies/implies?

vikasverma1077 commented 5 years ago

It does not imply anything during training. You should look at the training loss. We can not use the reweighted targets because for computing the accuracy we need discrete targets.

On Mon, Sep 9, 2019, 6:31 AM Sai Kalyan Siddanatham < notifications@github.com> wrote:

I was trying to implement mix-up-hidden for preactresnet18, I found that for calculating accuracy you are comparing mix-up output with original target instead of re-weighted one, that makes accuracy low during training, I didn't understand what that accuracy signifies/implies?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/vikasverma1077/manifold_mixup/issues/9?email_source=notifications&email_token=ADJ5QZQ3DGOXP3C4J4NC243QIYQZZA5CNFSM4IUZOW52YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HKEJF7A, or mute the thread https://github.com/notifications/unsubscribe-auth/ADJ5QZQKCJEY6CQP32VSS6DQIYQZZANCNFSM4IUZOW5Q .