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 2021/05/21 08:08:44 UTC

[GitHub] [tvm] ZhuYuJin opened a new issue #8104: Run slower than original pytorch?

ZhuYuJin opened a new issue #8104:
URL: https://github.com/apache/tvm/issues/8104


   My mac use the following process:
   https://ark.intel.com/content/www/us/en/ark/products/134906/intel-core-i7-8750h-processor-9m-cache-up-to-4-10-ghz.html
   
   Run the official demo. It seems TVM(cost 28ms) run slower than pytorch(cost 22ms), and much slower than onnxruntime(cost 7ms).
   
   I also find a blog that has the same situation with me...
   https://zhuanlan.zhihu.com/p/88369758
   
   ![image](https://user-images.githubusercontent.com/5879410/119104127-7b132600-ba4e-11eb-812a-4e1c887fd747.png)
   
   
   


-- 
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

[GitHub] [tvm] tqchen commented on issue #8104: Run slower than original pytorch?

Posted by GitBox <gi...@apache.org>.
tqchen commented on issue #8104:
URL: https://github.com/apache/tvm/issues/8104#issuecomment-845943299


   Thanks for asking the question, please open a new thread on https://discuss.tvm.apache.org/, this likely have things to do with the tuning being involved. https://github.com/tlc-pack/tlcbench also contains some more benchmark scripts that might be relevant


-- 
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

[GitHub] [tvm] tqchen closed issue #8104: Run slower than original pytorch?

Posted by GitBox <gi...@apache.org>.
tqchen closed issue #8104:
URL: https://github.com/apache/tvm/issues/8104


   


-- 
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