This repository has been archived by the owner on Nov 17, 2023. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 6.8k
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Yiyan66
requested review from
anirudh2290,
eric-haibin-lin and
szha
as code owners
February 16, 2020 13:43
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could u add a summary of what are changed in this PR?
@eric-haibin-lin Description part updated. |
reminisce
reviewed
Feb 17, 2020
Yiyan66
force-pushed
the
fallback_test
branch
from
February 17, 2020 07:24
901609d
to
b9af070
Compare
Yiyan66
force-pushed
the
fallback_test
branch
from
February 18, 2020 07:51
5db2332
to
a36fd2a
Compare
zheyuye
pushed a commit
to zheyuye/incubator-mxnet
that referenced
this pull request
Feb 19, 2020
* Add fallback ops * More fallback ops * Add more fallback ops * Fix astype * Support mixed type of mx.np.ndarray and np.ndarray in binary ops * More fix * Add result_type * Remove implemented ops * Fix * Add more tests * fallback--------------- * 8 * all fallback * ok * delete repetition Co-authored-by: reminisce <[email protected]>
anirudh2290
pushed a commit
to anirudh2290/mxnet
that referenced
this pull request
May 29, 2020
* Add fallback ops * More fallback ops * Add more fallback ops * Fix astype * Support mixed type of mx.np.ndarray and np.ndarray in binary ops * More fix * Add result_type * Remove implemented ops * Fix * Add more tests * fallback--------------- * 8 * all fallback * ok * delete repetition Co-authored-by: reminisce <[email protected]>
@Yiyan66 Did you intend to leave this line in the PR, as it prints whenever the file is imported: |
I suggest that be removed.. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This PR implements a mechanism of easily falling back to the official NumPy if operators do not have the corresponding native implementations in MXNet. This allows us to quickly have a decent coverage of NumPy operators in order to test MXNet as a backend compute engine for other machine learning libraries heavily depending on NumPy, such as sklearn, while adding more operators with native implementations in parallel.
The mechanism implemented in this PR only supports calling operators in the imperative way without autograd, i.e., through the module
mxnet.numpy
. More sophisticated fallback mechanism for supporting symbolic paradigm and autograd should resort to #16923.Checklist
Essentials
Please feel free to remove inapplicable items for your PR.
Changes
Comments