[Dynamo][Better Engineering] Type coverage for `torch/_dynamo/utils.py` #159580

Lucaskabela · 2025-07-31T17:20:02Z

As part of better engineering effort, we would like to improve out type support to improve dev experience in dynamo

This PR adds strict typing support to torch/_dynamo/utils.py

Running

mypy torch/_dynamo/utils.py --linecount-report /tmp/coverage_log

--------	Lines Annotated	Lines Total	% lines covered	Funcs Annotated	Funcs Total	% funcs covered
Main	2163	4792	45.14%	121	268	45.15%
This PR	4818	4818	100.00%	268	268	100.00%
Delta	+2655	+26	+54.84%	+147	0	+54.85%

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames

pytorch-bot · 2025-07-31T17:20:05Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159580

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (1 Unrelated Failure)

As of commit 1187733 with merge base 2a286cb ():

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

pull / linux-jammy-py3_9-clang9-xla / test (xla, 1, 1, lf.linux.12xlarge, unstable) (gh) (#158876)
sccache: error: couldn't connect to server

This comment was automatically generated by Dr. CI and updates every 15 minutes.

Lucaskabela · 2025-07-31T17:21:19Z

@pytorchbot label "topic: not user facing"

torch/_dynamo/utils.py

Lucaskabela · 2025-08-04T18:55:12Z

@pytorchmergebot merge

pytorchmergebot · 2025-08-04T18:57:02Z

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here

Skylion007

Some nits about these types

Skylion007 · 2025-08-04T15:53:39Z

torch/_dynamo/replay_record.py

    code_options: dict[str, Any] = field(default_factory=dict)

-    def dump(self, f: IO[str]) -> None:
+    def dump(self, f: Union[IO[str], BufferedWriter]) -> None:


Interesting, bufferedWriter doesn't necessarily subclass this? This is an fsspec bug, isn't it. :(

Skylion007 · 2025-08-04T15:53:40Z

torch/_dynamo/replay_record.py

    code_options: dict[str, Any] = field(default_factory=dict)

-    def dump(self, f: IO[str]) -> None:
+    def dump(self, f: Union[IO[str], BufferedWriter]) -> None:


Interesting, bufferedWriter doesn't necessarily subclass this? This is an fsspec bug, isn't it. :(

Skylion007 · 2025-08-04T15:55:16Z

torch/_dynamo/utils.py



-def hashable(x):
+def hashable(x: Any) -> bool:


Use a typing_extension.TypeIs[Hashable] as return type. Also this is very antiquated, you can use the abc protocol for this now, can't you?

Skylion007 · 2025-08-04T15:56:09Z

torch/_dynamo/utils.py



-def is_numpy_float_type(value):
+def is_numpy_float_type(value: Any) -> bool:


This could be TypeIs/Typeguard

These numpy types may be undefined unfortunately if the import fails, so I am not sure how to use them as type annotations in this event :/

Skylion007 · 2025-08-04T15:56:25Z

torch/_dynamo/utils.py



-def is_numpy_int_type(value):
+def is_numpy_int_type(value: Any) -> bool:


Same here. Allows type narrowing

torch/_dynamo/utils.py

Skylion007 · 2025-08-04T16:09:17Z

torch/_dynamo/utils.py

    return fn(*args, **kwargs)


-def is_utils_checkpoint(obj):


Could be TypeIs return type with TypeChecking block to avoid circular import

Skylion007 · 2025-08-04T20:47:40Z

@Lucaskabela Feel free to do a subsequent PR to further improve these

Lucaskabela · 2025-08-04T21:04:36Z

@Skylion007 thanks for the feedback! I will follow up shortly with improvements :) Honestly before these comments I was not very familiar with TypeIs/TypeGuard for narrowing so thank you for bringing these to my attention!

Summary: As part of better engineering effort, we would like to improve out type support to improve dev experience in dynamo This PR adds strict typing support to `torch/_dynamo/utils.py` Running ``` mypy torch/_dynamo/utils.py --linecount-report /tmp/coverage_log ``` | -------- | Lines Annotated | Lines Total | % lines covered | Funcs Annotated | Funcs Total | % funcs covered | | -------- | ------- | -------- | ------- | ------- | ------- | ------- | | Main | 2163 | 4792 | 45.14% | 121 | 268 | 45.15% | | This PR | 4818 | 4818 | 100.00% | 268 | 268 | 100.00% | | Delta | +2655 | +26 | +54.84% | +147 | 0 | +54.85% | X-link: pytorch/pytorch#159580 Approved by: https://github.com/williamwen42 Reviewed By: izaitsevfb Differential Revision: D79597355 fbshipit-source-id: 5bb7a21d68ef3c24659c5a79e148caeb9219a11f

Follow up to #159580 Pull Request resolved: #159824 Approved by: https://github.com/williamwen42

…h#159824) Follow up to pytorch#159580 Pull Request resolved: pytorch#159824 Approved by: https://github.com/williamwen42

…y` (pytorch#159580) As part of better engineering effort, we would like to improve out type support to improve dev experience in dynamo This PR adds strict typing support to `torch/_dynamo/utils.py` Running ``` mypy torch/_dynamo/utils.py --linecount-report /tmp/coverage_log ``` | -------- | Lines Annotated | Lines Total | % lines covered | Funcs Annotated | Funcs Total | % funcs covered | | -------- | ------- | -------- | ------- | ------- | ------- | ------- | | Main | 2163 | 4792 | 45.14% | 121 | 268 | 45.15% | | This PR | 4818 | 4818 | 100.00% | 268 | 268 | 100.00% | | Delta | +2655 | +26 | +54.84% | +147 | 0 | +54.85% | Pull Request resolved: pytorch#159580 Approved by: https://github.com/williamwen42

…h#159824) Follow up to pytorch#159580 Pull Request resolved: pytorch#159824 Approved by: https://github.com/williamwen42

pytorch-bot bot added ciflow/inductor module: dynamo labels Jul 31, 2025

pytorch-bot bot added the topic: not user facing topic category label Jul 31, 2025

Lucaskabela changed the title ~~Lucaskabela/typing utils.py~~ [Dynamo][Better Engineering] Type coverage for torch/_dynamo/utils.py Jul 31, 2025

Skylion007 reviewed Jul 31, 2025

View reviewed changes

torch/_dynamo/utils.py Outdated Show resolved Hide resolved

torch/_dynamo/utils.py Outdated Show resolved Hide resolved

torch/_dynamo/utils.py Outdated Show resolved Hide resolved

Lucaskabela force-pushed the lucaskabela/typing_utils.py branch from 1eb77e6 to a6dc480 Compare July 31, 2025 21:36

Lucaskabela requested review from Raymo111, StrongerXi and williamwen42 August 1, 2025 15:54

Lucaskabela marked this pull request as ready for review August 1, 2025 15:54

Lucaskabela added 5 commits August 1, 2025 09:00

Type torch/dynamo/utils.py

94608b2

Add type coverage for utils.py

bfc2f3a

fix signals

e9c87f7

Address comments

e95fca3

Fix np -> tnp

1187733

Lucaskabela force-pushed the lucaskabela/typing_utils.py branch from d58c817 to 1187733 Compare August 1, 2025 16:00

Lucaskabela requested a review from Skylion007 August 1, 2025 16:19

williamwen42 approved these changes Aug 1, 2025

View reviewed changes

pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Aug 4, 2025

pytorchmergebot added the merging label Aug 4, 2025

Skylion007 reviewed Aug 4, 2025

View reviewed changes

pytorchmergebot closed this in a7f3bdf Aug 4, 2025

pytorchmergebot added the Merged label Aug 4, 2025

pytorchmergebot removed the merging label Aug 4, 2025

Lucaskabela mentioned this pull request Aug 6, 2025

[BE][Dynamo] Type improvements in _dynamo/utils to generics #159824

Closed

github-actions bot deleted the lucaskabela/typing_utils.py branch September 4, 2025 02:07



		def is_numpy_float_type(value):
		def is_numpy_float_type(value: Any) -> bool:



		def is_numpy_int_type(value):
		def is_numpy_int_type(value: Any) -> bool:

[Dynamo][Better Engineering] Type coverage for torch/_dynamo/utils.py #159580

[Dynamo][Better Engineering] Type coverage for torch/_dynamo/utils.py #159580

Uh oh!

Conversation

Lucaskabela commented Jul 31, 2025 • edited by pytorch-bot bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Jul 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159580

✅ You can merge normally! (1 Unrelated Failure)

Uh oh!

Lucaskabela commented Jul 31, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Lucaskabela commented Aug 4, 2025

Uh oh!

pytorchmergebot commented Aug 4, 2025

Merge started

Uh oh!

Skylion007 left a comment

Choose a reason for hiding this comment

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Lucaskabela Aug 5, 2025

Choose a reason for hiding this comment

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Skylion007 Aug 4, 2025

Choose a reason for hiding this comment

Uh oh!

Skylion007 commented Aug 4, 2025

Uh oh!

Lucaskabela commented Aug 4, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

[Dynamo][Better Engineering] Type coverage for `torch/_dynamo/utils.py` #159580

[Dynamo][Better Engineering] Type coverage for `torch/_dynamo/utils.py` #159580

Lucaskabela commented Jul 31, 2025 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Jul 31, 2025 •

edited

Loading