Allow partitioning quantized linear for FP32-only partition #7284
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7284
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Cancelled JobAs of commit 9b1bb48 with merge base 9d1a310 ( NEW FAILURE - The following job has failed:
CANCELLED JOB - The following job was cancelled. Please retry:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
This PR needs a
|
digantdesai
added a commit
to digantdesai/executorch-1
that referenced
this pull request
Dec 11, 2024
…7284) Summary: Pull Request resolved: pytorch#7284 Add overwrite precision of linear op in partitioning. When using legacy_mode, we will test we don't partition [add]mm given, (1) We can't assume that weights are always static (non param). (2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias. (2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI. Added tests to make sure we see this behavior Differential Revision: D67011716
digantdesai
force-pushed
the
export-D67011716
branch
from
2ad1d6c to
7b32bc5
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
digantdesai
added a commit
to digantdesai/executorch-1
that referenced
this pull request
Dec 11, 2024
…7284) Summary: Pull Request resolved: pytorch#7284 Add overwrite precision of linear op in partitioning. When using legacy_mode, we will test we don't partition [add]mm given, (1) We can't assume that weights are always static (non param). (2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias. (2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI. Added tests to make sure we see this behavior Differential Revision: D67011716
digantdesai
force-pushed
the
export-D67011716
branch
from
7b32bc5 to
ed167b6
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
digantdesai
force-pushed
the
export-D67011716
branch
from
ed167b6 to
adf9fc1
Compare
digantdesai
added a commit
to digantdesai/executorch-1
that referenced
this pull request
Dec 17, 2024
…7284) Summary: Add overwrite precision of linear op in partitioning. When using legacy_mode, we will test we don't partition [add]mm given, (1) We can't assume that weights are always static (non param). (2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias. (2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI. Added tests to make sure we see this behavior Reviewed By: mcr229 Differential Revision: D67011716
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
digantdesai
added
the
module: xnnpack
digantdesai
added a commit
to digantdesai/executorch-1
that referenced
this pull request
Dec 17, 2024
…7284) Summary: Add overwrite precision of linear op in partitioning. When using legacy_mode, we will test we don't partition [add]mm given, (1) We can't assume that weights are always static (non param). (2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias. (2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI. Added tests to make sure we see this behavior Reviewed By: mcr229 Differential Revision: D67011716
digantdesai
force-pushed
the
export-D67011716
branch
from
adf9fc1 to
8b47c4d
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
…7284) Summary: Add overwrite precision of linear op in partitioning. When using legacy_mode, we will test we don't partition [add]mm given, (1) We can't assume that weights are always static (non param). (2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias. (2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI. Added tests to make sure we see this behavior Reviewed By: mcr229 Differential Revision: D67011716
digantdesai
force-pushed
the
export-D67011716
branch
from
8b47c4d to
9b1bb48
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D67011716 |
mcr229
approved these changes
Dec 18, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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:
Add overwrite precision of linear op in partitioning.
When using legacy_mode, we will test we don't partition [add]mm given,
(1) We can't assume that weights are always static (non param).
(2) Alternatively, when lowering [add]mm to xnn::bmm we can't support bias.
(2)(a) Only lowering non-bias [add]mm, which is only exposed on legacy_path deemed low ROI.
Added tests to make sure we see this behavior
Differential Revision: D67011716