Allow disabling int specialization on nn.Modules #142829
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/142829
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit 64924d1 with merge base bd7d81d ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D66837042 |
jhadidjojo
force-pushed
the
export-D66837042
branch
from
6c40a3c to
b97be90
Compare
|
@pytorchbot label "release notes: dynamo" |
|
This pull request was exported from Phabricator. Differential Revision: D66837042 |
jhadidjojo
force-pushed
the
export-D66837042
branch
from
b97be90 to
03c342b
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66837042 |
|
@pytorchbot merge |
|
This PR needs to be approved by an authorized maintainer before merge. |
Summary: Resolves issue [pytorch#140464](pytorch#140464) by adding an option to not specialize int from nn.Modules (False by default to maintain existing behavior). Test Plan: `buck2 test mode/opt caffe2/test/dynamo:test_dynamo -- test_modules.py::NNModuleTests::test_nn_module_unspec_int_attr` Also stepped through the repro in [pytorch#140464](pytorch#140464) to confirm that, with the flag set to True, the int is no longer specialized. Reviewed By: ezyang, yanboliang Differential Revision: D66837042
jhadidjojo
force-pushed
the
export-D66837042
branch
from
03c342b to
64924d1
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66837042 |
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 1 mandatory check(s) failed. The first few are: Dig deeper by viewing the failures on hud |
|
@pytorchbot merge -f "Failures unrelated" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
5 participants
Resolves issue #140464 by adding an option to not specialize int from nn.Modules (False by default to maintain existing behavior).
Test Plan:
buck2 test mode/opt caffe2/test/dynamo:test_dynamo -- test_modules.py::NNModuleTests::test_nn_module_unspec_int_attrDifferential Revision: D66837042
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames