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 2019/11/12 00:05:31 UTC
[GitHub] [incubator-tvm] zxy844288792 opened a new pull request #4310:
[Relay][Frontend][Keras] batch_norm op params not handling well
zxy844288792 opened a new pull request #4310: [Relay][Frontend][Keras] batch_norm op params not handling well
URL: https://github.com/apache/incubator-tvm/pull/4310
I have a user case shown below that a batch norm op with attributes center=True, scale=False. Thus there is no gamma in the model so 'gamma' will not be defined in params. However, in op.nn.batch_norm, both beta and gamma are required. In the case, we should detect if beta/gamma exists and initilize them if not.
Thanks for contributing to TVM! Please refer to guideline https://docs.tvm.ai/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from [Reviewers](https://github.com/apache/incubator-tvm/blob/master/CONTRIBUTORS.md#reviewers) by @ them in the pull request thread.
----------------------------------------------------------------
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