How to use torch-directml in transformer's AutoModel and AutoTokenizer?

[fenixlam]

As far as I know, some of the functions can be used with torch-directml.
But I didn't find any examples about Transformers's AutoModel and AutoTokenizer calling with torch-directml.
I want to ask if they can use directml GPU?

[LasonHistory]

it seems automodel dont recognize AMD-GPU, the error is
File "C:\Users\IntoTheWorld/.cache\huggingface\modules\transformers_modules\models\quantization.py", line 464, in quantize
assert kernels is not None
AssertionError

[Adele101]

Hello, thank you for submitting this issue. While I can't provide a timeline for resolution as the moment, please know that your feedback is valuable to us. We will follow up once we can review this issue.