Arm backend: Insert transposes between <4D and 4D tensors #6045
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/6045
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 88ee974 with merge base b1c94ab ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
oscarandersson8218
added
partner: arm
facebook-github-bot
added
the
CLA Signed
oscarandersson8218
force-pushed
the
transpose_insertion
branch
2 times, most recently
from
27be916 to
927f143
Compare
|
@digantdesai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
digantdesai
approved these changes
Oct 17, 2024
Tensors of rank < 4 always have a contiguous dim_order while 4D tensors may have (0, 2, 3, 1) as dim_order. This patch inserts transposes between 4D tensors and other tensors, more precisely it targets squeeze and unsqueeze ops. The transposes will act as bridges between different dim_orders to ensure that activations are passed forward in the correct format. Vela version is bumped to include transpose implementation. This version also includes implementation of transpose, therefore xfails in test_expand and test_repeat are disabled. Signed-off-by: Oscar Andersson <[email protected]> Change-Id: If56344b2153441f113e4e87e95d41acad5df9220
oscarandersson8218
force-pushed
the
transpose_insertion
branch
from
927f143 to
88ee974
Compare
|
@digantdesai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
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.
Tensors of rank < 4 always have a contiguous dim_orde (e.g. (0, 1, 2)) while 4D tensors always have (0, 2, 3, 1) as dim_order. This patch inserts transposes between 4D tensors and lower ranked tensors, more precisely it targets squeeze and unsqueeze ops. The transposes will act as bridges between different dim_orders to ensure that activations are passed forward in the correct format.
Vela version is bumped to include transpose implementation. This version also includes implementation of transpose, therefore xfails in test_expand and test_repeat are disabled.