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 2018/06/15 16:50:16 UTC

[GitHub] leezu commented on issue #11271: nd.topk regression with nan values

leezu commented on issue #11271: nd.topk regression with nan values
URL: https://github.com/apache/incubator-mxnet/issues/11271#issuecomment-397680342
 
 
   I am not aware of any documentation. Maybe @szha has more insights.
   
   Already in v1.2 the behavior of `topk` vs `argmax` was inconsistent under the presence of `nan`. However, both returned meaningful results under the respective assumption to to treat `nan` as top value or to ignore `nan` values. Currently `topk` treats the whole order of the array as being undefined if it contains any `nan` value. It would be better to adopt a treatment of `nan` such that the order of the array to still well defined.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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