Cross Entropy loss or NLL loss

PRIS-CV / Fine-Grained-or-Not

Code release for Your “Flamingo” is My “Bird”: Fine-Grained, or Not (CVPR 2021 Oral)

MIT License

58 stars 9 forks source link

Cross Entropy loss or NLL loss #1

Closed 07Agarg closed 3 years ago

07Agarg commented 3 years ago

Hi, Thanks for the nice repo!

I just wanted clarification about the NLL criterion in the code. Are the results in Birds.out using Cross Entropy criterion or NLL criterion?

Also, I tried running the code as is (using NLL criterion), train and test loss becomes nan.

Thanks once again.

dongliangchang commented 3 years ago

The results in Birds.out are using criterion_NLLLoss. And I re-run the code in this repo, all thing is fine. Please check your environment and the data preparation.

07Agarg commented 3 years ago

Thank you!