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 2019/07/09 08:47:23 UTC
[GitHub] [incubator-mxnet] zoeygxy opened a new pull request #15494: [Numpy]
Added operator logaddexp2;
added support for zero-size tensor in BinaryBroadcastBackwardUseIn
zoeygxy opened a new pull request #15494: [Numpy] Added operator logaddexp2; added support for zero-size tensor in BinaryBroadcastBackwardUseIn
URL: https://github.com/apache/incubator-mxnet/pull/15494
## Description ##
- Added [np.logaddexp2](https://docs.scipy.org/doc/numpy/reference/generated/numpy.logaddexp2.html) and its docs for future rendering. Scalar calculation is supported.
- Fixed zero-size tensor error in BinaryBroadcastBackwardUseIn (GPU version).
## Checklist ##
### Essentials ###
- [x] Changes are complete (i.e. I finished coding on this PR)
- [x] All changes have test coverage:
- Unit tests are added for small changes to verify correctness
- [x] Code is well-documented:
- For user-facing API changes, API doc string has been updated.
- [x] To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
### Changes ###
- [x] Added [np.logaddexp2](https://docs.scipy.org/doc/numpy/reference/generated/numpy.logaddexp2.html) and its docs for future rendering. Scalar calculation is supported.
- [x] Fixed zero-size tensor error in BinaryBroadcastBackwardUseIn.
## Comments ##
- The previous error case description: when one operand is non-zero-size (eg. shape of (1, 3)) and the other is zero-size (eg. shape of (2, 0, 3)), the backward calculation on GPU will raise a floating point error.
- Please merge this PR before merging implementation for np.logaddexp, the similar operator that also depends on the BinaryBroadcastBackwardUseIn fix.
----------------------------------------------------------------
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