What is the proper range of NT xent loss when the network converges?

sthalles / SimCLR

PyTorch implementation of SimCLR: A Simple Framework for Contrastive Learning of Visual Representations

https://sthalles.github.io/simple-self-supervised-learning/

MIT License

2.19k stars 457 forks source link

What is the proper range of NT xent loss when the network converges? #21

Closed ckLibra closed 3 years ago

ckLibra commented 3 years ago

After 80 epoch,my nt xent loss is about 5.I am not sure whether the network has learnt well.

sthalles commented 3 years ago

Hello guys, that is a general message to say that I have refactored the whole project. Please I a look at the new impl and free to submit PR if you find any bugs. Thanks.

muyuuuu commented 2 years ago

After 80 epoch,my nt xent loss is about 5.I am not sure whether the network has learnt well.

I'm also meeting this problem, is it normal ? or how do you fix it?