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

Posted to issues@mxnet.apache.org by "Junru Shao (JIRA)" <ji...@apache.org> on 2018/07/17 21:22:00 UTC

[jira] [Created] (MXNET-684) Add `ifelse`

Junru Shao created MXNET-684:
--------------------------------

             Summary: Add `ifelse`
                 Key: MXNET-684
                 URL: https://issues.apache.org/jira/browse/MXNET-684
             Project: Apache MXNet
          Issue Type: Improvement
            Reporter: Junru Shao


This PR adds a control flow operator: ifelse. It takes a Python function as input and run the function over the elements in the input array. ifelse is similar to tf.cond in TensorFlow.

This PR is part of the proposal of adding a set of control flow operators to MXNet.
 [https://cwiki.apache.org/confluence/display/MXNET/Optimize+dynamic+neural+network+models+with+control+flow+operators]

See also MXNET-432 and MXNET-626.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org