helsenIgnace comments

Repositories
Issues
Comments

Results 2 comments of


                                            helsenIgnace

RuntimeError: mat1 and mat2 must have the same dtype, but got Float and BFloat16

I had the same error, turned out an 'with torch.cuda.amp.autocast(dtype=torch.float32)'. Be sure to check whether you changed all the torch.cuda.amp.autocast() to use torch.float32 instead of bfloat16.

[MPS] Add support for autocast in MPS

When do you think this pr could be merged?