🦊 [AI Accelerators] Consolidate native_layer_norm for nested tensor #86295
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/86295
Note: Links to docs will display an error until the docs builds have been completed. ✅ No Failures, 7 PendingAs of commit e536f69: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
|
|
This pull request was exported from Phabricator. Differential Revision: D40105207 |
ani300
force-pushed
the
export-D40105207
branch
from
f354bbd to
d7e89f9
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40105207 |
…ytorch#86295) Summary: Pull Request resolved: pytorch#86295 In order to make the layer normalization implementation for nested tensors public, it needs to be generalized to accept a normalized_shape argument instead of assuming it to be the last dimension of the nested_tensor. This commit does that, as well as adding extra unit tests to ensure the implementation is correct. Test Plan: All unit tests designed to test different ways of using the function work: `buck test //caffe2/test:nested -- test_layer_norm` Differential Revision: D40105207 fbshipit-source-id: fe6b9c602aaf1a115101decb384f979bf02cb224
ani300
force-pushed
the
export-D40105207
branch
from
d7e89f9 to
e536f69
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40105207 |
drisspg
added
module: nestedtensor
|
@pytorchbot merge |
|
@pytorchbot successfully started a merge job. Check the current status here. |
|
Hey @ani300. |
…?e=20native=5Flayer=5Fnorm=20for=20nested=20tensor=20(#86295)?= (#86295) Summary: In order to make the layer normalization implementation for nested tensors public, it needs to be generalized to accept a normalized_shape argument instead of assuming it to be the last dimension of the nested_tensor. This commit does that, as well as adding extra unit tests to ensure the implementation is correct. Pull Request resolved: #86295 Approved by: https://github.com/drisspg Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/cdbffa7f665dd144ada92c11d223aeb8b5c3887a Test plan from GitHub: All unit tests designed to test different ways of using the function work: `buck test //caffe2/test:nested -- test_layer_norm` Original Phabricator Test Plan: All unit tests designed to test different ways of using the function work: `buck test //caffe2/test:nested -- test_layer_norm` Reviewed By: drisspg, seemethere Differential Revision: D40105207 fbshipit-source-id: a54ee614888411a9bdd0f57ce9c11efb5d61467b
4 participants
Summary: In order to make the layer normalization implementation for nested tensors public, it needs to be generalized to accept a normalized_shape argument instead of assuming it to be the last dimension of the nested_tensor. This commit does that, as well as adding extra unit tests to ensure the implementation is correct.
Test Plan:
All unit tests designed to test different ways of using the function work:
buck test //caffe2/test:nested -- test_layer_normDifferential Revision: D40105207