[ET-VK] Change weight packing in embedding #7063
Merged
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
The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed. However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing. This diff changes the implementation to use height-packing. Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7063
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit f958cab with merge base fbcc9a1 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
yipjustin
added a commit
that referenced
this pull request
Nov 25, 2024
The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed. However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing. This diff changes the implementation to use height-packing. Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/) ghstack-source-id: 255375459 Pull Request resolved: #7063
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66421366 |
junpi3
approved these changes
Nov 25, 2024
nathanaelsee
approved these changes
Nov 25, 2024
The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed. However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing. This diff changes the implementation to use height-packing. Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/) [ghstack-poisoned]
yipjustin
added a commit
that referenced
this pull request
Nov 26, 2024
Pull Request resolved: #7063 The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed. However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing. This diff changes the implementation to use height-packing. ghstack-source-id: 255439082 Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/)
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66421366 |
kirklandsign
added a commit
that referenced
this pull request
Nov 26, 2024
Pull Request resolved: #7063 The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed. However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing. This diff changes the implementation to use height-packing. ghstack-source-id: 255439082 Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/) Co-authored-by: Justin Yip <[email protected]>
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 (oldest at bottom):
The existing weight tensor for aten.embedding is created using a
tensor_likefrom the output tensor, which defaults to channel packed.However, the weight tensor is actually a 2D-tensor of
(num_embedding, dim_of_embedding). It is better in space to use either width or height packing.This diff changes the implementation to use height-packing.
Differential Revision: D66421366