Integrating torchvision models and Lightning

[vasudev13]

Hi, awesome work on AUC maximization. I'm trying to reproduce the second step from https://arxiv.org/abs/2012.03173 , I sort of have a different pre-training approach.

However, I'm using a torchvision.models.resnet instead of models from libauc, it appears it should fairly easy to use the former when replicated the mentioned tutorial for CheXpert. (Essentially is libauc.models.ResNet18 replaceable with torchvision.models.resnet18 ?)

Also, I'm using libraries like pytorch-lightning and wandb, do you expect any possible irregularities I might face if I use them with LibAUC?

Thanks for your time and amazing work!

[yzhuoning]

Thanks! It should work fine with torchvision models, but just remember to check if you need to include an activation function after model outputs. I haven't checked compatibility with pytorch-lightning yet, but I plan to do it in near future.

[vasudev13]

Awesome, thank you so much. Just to be sure you mean use something like toch.sigmoid over model outputs for binary and multi-label predictions right?

I'll circle back incase I need some more inputs on usage.
Thanks

[RoyBenjamin]

@vasudev13 I am wondering if you managed to use LibAUC with pytorch-lightning as I am experiencing this issue:
TypeError: step() got an unexpected keyword argument 'closure'
When using PESG.