[pt] Add incude_last_offset option to EmbeddingBag mean and max #42215
Conversation
This file contains hidden or 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
Specifically on #27477 (comment) We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/) Differential Revision: [D22801881](https://our.internmc.facebook.com/intern/diff/D22801881/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D22801881/)! [ghstack-poisoned]
jianyuh
requested review from
apaszke,
ebetica,
goldsborough and
yf225
as code owners
💊 CI failures summary and remediationsAs of commit daacef0 (more details on the Dr. CI page):
🕵️ 3 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
dzhulgakov
approved these changes
Jul 29, 2020
Collaborator
dzhulgakov
left a comment
dzhulgakov
left a comment
There was a problem hiding this comment.
Looking good, just fix the test a bit please! And lint seems to be complaining
test/test_nn.py
Outdated
| for dtype, mode, trainable, include_last_offset in itertools.product(dtypes, modes, trainable_scale, include_last_offset): | ||
| test_per_sample_weights_new_offsets(mode, dtype, trainable, include_last_offset) | ||
| if mode != 'sum': | ||
| has_weight = False |
Collaborator
There was a problem hiding this comment.
this actually disables has_weight for future iterations of the loop too :(
just pass has_weight and mode == 'sum' below as an arg
test/test_nn.py
Outdated
| modes = ('sum', 'max', 'mean') | ||
| trainable_scale = (True, False) | ||
| include_last_offset = (True, False) | ||
| has_weight = (True, False) |
Collaborator
There was a problem hiding this comment.
you probably wanted to include has_weight in the for loop iteration too
or you could do for simplicity:
modes = (('sum', False), ('sum', True), ('max', False), ('mean', False))
for dtype, (mode, has_weight), trainable, ...
test/test_nn.py
Outdated
| offsets = torch.tensor([0, 0, 3, 3, 6], device=device, dtype=torch.long) | ||
|
|
||
| if include_last_offset is True and mode == 'sum': | ||
| if include_last_offset is True: |
Collaborator
There was a problem hiding this comment.
nit: no need for is True :)
houseroad
reviewed
Jul 29, 2020
test/test_nn.py
Outdated
| def _embedding_bag_reference_impl(self, input, weight, offsets=None, mode='sum', | ||
| per_sample_weights=None, include_last_offset=False): | ||
| assert mode == 'sum' | ||
| assert mode == 'sum' or per_sample_weights == None |
…d max" Specifically on #27477 (comment) We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/) Differential Revision: [D22801881](https://our.internmc.facebook.com/intern/diff/D22801881/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D22801881/)! [ghstack-poisoned]
…d max" Specifically on #27477 (comment) We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/) Differential Revision: [D22801881](https://our.internmc.facebook.com/intern/diff/D22801881/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D22801881/)! [ghstack-poisoned]
…d max" Specifically on #27477 (comment) We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/) Differential Revision: [D22801881](https://our.internmc.facebook.com/intern/diff/D22801881/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D22801881/)! [ghstack-poisoned]
jianyuh
added a commit
that referenced
this pull request
Jul 29, 2020
Pull Request resolved: #42215 Specifically on #27477 (comment) We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/). More details: https://www.internalfb.com/intern/diff/D22794469/?dest_fbid=309597093427021&transaction_id=631457624153457 ghstack-source-id: 108733009 Differential Revision: [D22801881](https://our.internmc.facebook.com/intern/diff/D22801881/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D22801881/)!
Contributor
|
This pull request has been merged in 1c5c289. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Stack from ghstack:
Specifically on #27477 (comment)
We would like to supported with include_last=True overall for other reduction types like mean and max. It now causes further code fragmentation in DPER (https://www.internalfb.com/intern/diff/D22794469/)
Differential Revision: D22801881
NOTE FOR REVIEWERS: This PR has internal Facebook specific changes or comments, please review them on Phabricator!