Quan Vuong

Results 16 comments of Quan Vuong

I create a self-contained snippet that illustrates the issue. Thank you for your attention!

Sorry about closing and opening the issue (misclick). I'm happy to send a PR in. It's literally 2 lines : ).

adding del loss and torch.cuda.empty_cache() solve this problem

Actually, using empty_cached() leads to really slow GPU operations (60 hours for the fine tuning step). Is there another work around? If I simply do del loss without emptying the...

you can add it after loss.backward()

Is there any update on when text-alignment will be supported?

Any chance there's an update on text-alignment? Thanks so much!

I installed Qt and [HFO](https://github.com/LARG/HFO)'s installation and sample examples ran fine ...