Shiv Kaul comments

Repositories
Issues
Comments

Results 3 comments of


                                            Shiv Kaul

I train controlnet_sdxl in bf16 datatype, got unsupported ERROR in datasets

I also see the same error and get passed it by casting that line to float. so `for x in obj.detach().cpu().numpy()` becomes `for x in obj.detach().to(torch.float).cpu().numpy()` I got the idea...

optimize seamless-m4t/vits model for text-to-speech generation

@sywangyi Can you please add CI tests

Skaulintel/ci mistral fp8

moved to https://github.com/huggingface/optimum-habana/pull/918