You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/10/12 23:14:34 UTC

[GitHub] [incubator-mxnet] Zha0q1 commented on pull request #19339: [WIP] Numpy Argmax(min) Rewrite

Zha0q1 commented on pull request #19339:
URL: https://github.com/apache/incubator-mxnet/pull/19339#issuecomment-707391025


   > Great.
   > Apart from the individual argmax kernel, can you talk about performance (+ve or -ve) on end to end customer use case?
   > Where is argmax used commonly?
   > Also, this will help make judgement call if 1.8x increase which 80% regression on existing behavior on GPU with small dim (<64).
   
   @sxjscience Would you suggest a model to test this operator on? 


----------------------------------------------------------------
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