You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2020/04/18 00:43:37 UTC
[GitHub] [incubator-tvm] masahi edited a comment on issue #5363: [Topi,
ARM] Disbale Winograd for quantized tensors.
masahi edited a comment on issue #5363: [Topi, ARM] Disbale Winograd for quantized tensors.
URL: https://github.com/apache/incubator-tvm/pull/5363#issuecomment-615523135
I will take a look at how MKLDNN does winograd on int8.
~~I also wonder if weight quantization is working correctly. Since transformed weight shape is not in (OC, OI, KH, KW) format, maybe per channel quantize is broken if winograd is not taken into acount.~~
UPDATE: Since weight quantization happens before winograd transform, it is not a problem.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services