support filling partial rows from backend #4158
Closed
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
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
This pull request was exported from Phabricator. Differential Revision: D75048595 |
duduyi2013
force-pushed
the
export-D75048595
branch
from
7d0a9b5 to
ab1e637
Compare
duduyi2013
added a commit
to duduyi2013/FBGEMM
that referenced
this pull request
May 21, 2025
Summary: X-link: facebookresearch/FBGEMM#1238 change set 1. After enabling optimizer offloading, when read optimizer out, we want to only read the tailing bytes out instead of the whole value field. This isn't supported in EmbeddingRocksdb before, add it here. 2. Once we have the functionality above, we dont need to call .narrow() or .continuous() on the output tensor anymore, remove those 3. We don't need optimizer chunking anymore(Ideally we can just return PMT to the checkpoint and let it handle KVT the same way as for weights) 4. fix UT bugs for constructing kv zch ssd tbe Differential Revision: D75048595
duduyi2013
added a commit
to duduyi2013/FBGEMM
that referenced
this pull request
May 21, 2025
Summary: X-link: facebookresearch/FBGEMM#1238 change set 1. After enabling optimizer offloading, when read optimizer out, we want to only read the tailing bytes out instead of the whole value field. This isn't supported in EmbeddingRocksdb before, add it here. 2. Once we have the functionality above, we dont need to call .narrow() or .continuous() on the output tensor anymore, remove those 3. We don't need optimizer chunking anymore(Ideally we can just return PMT to the checkpoint and let it handle KVT the same way as for weights) 4. fix UT bugs for constructing kv zch ssd tbe Differential Revision: D75048595
duduyi2013
force-pushed
the
export-D75048595
branch
from
ab1e637 to
99365a4
Compare
|
This pull request was exported from Phabricator. Differential Revision: D75048595 |
duduyi2013
added a commit
to duduyi2013/FBGEMM
that referenced
this pull request
May 21, 2025
Summary: Pull Request resolved: pytorch#4158 X-link: facebookresearch/FBGEMM#1238 change set 1. After enabling optimizer offloading, when read optimizer out, we want to only read the tailing bytes out instead of the whole value field. This isn't supported in EmbeddingRocksdb before, add it here. 2. Once we have the functionality above, we dont need to call .narrow() or .continuous() on the output tensor anymore, remove those 3. We don't need optimizer chunking anymore(Ideally we can just return PMT to the checkpoint and let it handle KVT the same way as for weights) 4. fix UT bugs for constructing kv zch ssd tbe Differential Revision: D75048595
duduyi2013
force-pushed
the
export-D75048595
branch
from
99365a4 to
298e192
Compare
|
This pull request was exported from Phabricator. Differential Revision: D75048595 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D75048595 |
Summary: Pull Request resolved: pytorch#4158 X-link: facebookresearch/FBGEMM#1238 change set 1. After enabling optimizer offloading, when read optimizer out, we want to only read the tailing bytes out instead of the whole value field. This isn't supported in EmbeddingRocksdb before, add it here. 2. Once we have the functionality above, we dont need to call .narrow() or .continuous() on the output tensor anymore, remove those 3. We don't need optimizer chunking anymore(Ideally we can just return PMT to the checkpoint and let it handle KVT the same way as for weights) 4. fix UT bugs for constructing kv zch ssd tbe Differential Revision: D75048595
|
This pull request was exported from Phabricator. Differential Revision: D75048595 |
duduyi2013
force-pushed
the
export-D75048595
branch
from
298e192 to
72411d4
Compare
|
This pull request has been merged in d68987c. |
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.
Summary:
change set
Differential Revision: D75048595