This repository has been archived by the owner on Mar 19, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 332
Tensor input to loss functions to they follow the pytorch definitions #79
Closed
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
Differential Revision: D22842068 fbshipit-source-id: 2fa1a38e4896cd6a96b24f8ea583cbf1a9186f6f
Differential Revision: D22856092 fbshipit-source-id: f630b1f312687850cb486a166e29c392344965cf
Differential Revision: D22865193 fbshipit-source-id: c504a5f20cee37251940a55e1ce2623e26c7f2ea
…nd key access Differential Revision: D22872266 fbshipit-source-id: 6c8c72bfa8d2289a4a89ecddb8c5b3421c0d911c
Differential Revision: D22874810 fbshipit-source-id: 194dd65f34c645c1f31582a30005a7e37e4d3280
Differential Revision: D22875254 fbshipit-source-id: 2461ca59726b50b3df8fd2457b633cbd19bc5340
Summary: in most loss functions in ssl, we operate on single tensors and the actual input is [tensor]. This is different from pytorch definitions of losses where the input is a tensor. fixing this in vissl to closely follow pytorch definitions. except `nce-loss` and `cross entropy loss with multiple outputs`, all losses now take the tensor as input Reviewed By: mannatsingh Differential Revision: D22877422 fbshipit-source-id: 124f2af17d9036b5aae3a1f5d7be13f11211113c
facebook-github-bot
added
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
fb-exported
labels
Aug 10, 2020
This pull request was exported from Phabricator. Differential Revision: D22877422 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Aug 10, 2020
…#79) Summary: Pull Request resolved: #79 in most loss functions in ssl, we operate on single tensors and the actual input is [tensor]. This is different from pytorch definitions of losses where the input is a tensor. fixing this in vissl to closely follow pytorch definitions. except `nce-loss` and `cross entropy loss with multiple outputs`, all losses now take the tensor as input Reviewed By: mannatsingh Differential Revision: D22877422 fbshipit-source-id: c4ef8d85f080d4034a8b308d1872e0ed56ceb75f
prigoyal
pushed a commit
to prigoyal/vissl
that referenced
this pull request
Apr 15, 2021
Summary: Pull Request resolved: fairinternal/ssl_scaling#79 This is a coagulate merge, coordinated by prigoyal and uhbuhb. It addresses the following issues/pull requests: 1. MLH-Fellowship#7 2. MLH-Fellowship#8 3. MLH-Fellowship#9 4. MLH-Fellowship#10 5. MLH-Fellowship#13 6. MLH-Fellowship#14 Pull Request resolved: facebookresearch#221 Differential Revision: D27608671 Pulled By: prigoyal fbshipit-source-id: d15232bb00d11db7b2c245a9312d3d5be165ada3 Co-authored-by: grace-omotoso
prigoyal
pushed a commit
to prigoyal/vissl
that referenced
this pull request
Apr 15, 2021
Summary: Pull Request resolved: fairinternal/ssl_scaling#79 This is a coagulate merge, coordinated by prigoyal and uhbuhb. It addresses the following issues/pull requests: 1. MLH-Fellowship#7 2. MLH-Fellowship#8 3. MLH-Fellowship#9 4. MLH-Fellowship#10 5. MLH-Fellowship#13 6. MLH-Fellowship#14 Pull Request resolved: facebookresearch#221 Differential Revision: D27608671 Pulled By: prigoyal fbshipit-source-id: ebe71ff5b25fa77394661d442062fdb127cc9a48 Co-authored-by: grace-omotoso
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Labels
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
fb-exported
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.
Summary:
in most loss functions in ssl, we operate on single tensors and the actual input is [tensor]. This is different from pytorch definitions of losses where the input is a tensor.
fixing this in vissl to closely follow pytorch definitions. except
nce-loss
andcross entropy loss with multiple outputs
, all losses now take the tensor as inputReviewed By: mannatsingh
Differential Revision: D22877422