-
Notifications
You must be signed in to change notification settings - Fork 568
Clean up WeightRow
in preparation for optimizer state offloading
#4021
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
This pull request was exported from Phabricator. Differential Revision: D73473546 |
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
Summary: - Replace if-statements with if-constexpr blocks to optimize out some code branches completely. This is to facilitate `WeightRow` class cleanups - Replace uses of WeightRow with WeightRowAccessor where cache loading and eviction are not used Differential Revision: D73678501
5afcdf0
to
bdf4726
Compare
This pull request was exported from Phabricator. Differential Revision: D73473546 |
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
2e7ddaf
to
174e1da
Compare
…ytorch#4021) Summary: X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
…ytorch#4021) Summary: X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
…ytorch#4021) Summary: X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
Summary: - Fix `int32_t` to `auto` for code around `WeightRow` - Fix `kINT8QparamsBytes` from `float` to `int32_t` Differential Revision: D73690651
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
Summary: - Simplify weight row cache load and evict routines Differential Revision: D73693209
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
…ytorch#4021) Summary: Pull Request resolved: pytorch#4021 X-link: facebookresearch/FBGEMM#1109 - Clean up `WeightRow` implementation in preparation for optimizer state offloading - Add documentation for the class Differential Revision: D73473546
This pull request was exported from Phabricator. Differential Revision: D73473546 |
Summary:
WeightRow
implementation in preparation for optimizer state offloadingDifferential Revision: D73473546