gz944367214 opened a new issue, #426:
URL: https://github.com/apache/tvm-ffi/issues/426
tvm-ffi version: 0.1.6
When I try to use flashinfer W4A8 FuseMoe Kernel, I pass the weight viewed
as torch.quint4x2, which cause an error:
File "python/tvm_ffi/cython/function.pxi", line 909, in
tvm_ffi.core.Function.__call__
File "python/tvm_ffi/cython/function.pxi", line 623, in
tvm_ffi.core.TVMFFIPyArgSetterDTypeFromTorch_
ValueError: ('Unsupported torch dtype: ', torch.quint4x2)
Are there any plans to support torch.quint4x2 in tvm-ffi?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
