Fabric: Enable auto for devices and accelerator cli` arguments
#20913
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
fnhirwa
requested review from
lantiga,
Borda,
tchaton,
justusschock and
ethanwharris
as code owners
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #20913 +/- ##
=======================================
- Coverage 87% 87% -0%
=======================================
Files 268 268
Lines 23442 23456 +14
=======================================
+ Hits 20394 20402 +8
- Misses 3048 3054 +6 |
Borda
reviewed
Jun 18, 2025
src/lightning/fabric/cli.py
Outdated
| @@ -187,6 +187,19 @@ def _set_env_variables(args: Namespace) -> None: | |||
|
|
|||
| def _get_num_processes(accelerator: str, devices: str) -> int: | |||
| """Parse the `devices` argument to determine how many processes need to be launched on the current machine.""" | |||
| if accelerator == "auto" or accelerator is None: | |||
There was a problem hiding this comment.
lets not duplicate the logic and use already written functions like:
Borda
reviewed
Jun 19, 2025
| _SUPPORTED_ACCELERATORS = ("cpu", "gpu", "cuda", "mps", "tpu", "auto") | ||
|
|
||
|
|
||
| def _choose_auto_accelerator() -> str: |
There was a problem hiding this comment.
so why do we have now the same function?
There was a problem hiding this comment.
I tried to import from
But there were some dependencies errors, one for
torchmetrics on some runners.
There was a problem hiding this comment.
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Outdated suggestions cannot be applied.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.
What does this PR do?
Fixes #20451
This PR adds support for the
autovariable foracceleratoranddevicesin the Fabric CLI.When
autois passed to--acceleratorwe now detect whether the machine has eithermps,cuda, orcpuand dynamically runs on the available device, this is the same case when no accelerator is specified.For
autoin the--devicesargument the devices default to1as specified in the docstring of the argument herepytorch-lightning/src/lightning/fabric/cli.py
Line 80 in c03660a
No breaking changes were introduced.
Before submitting
PR review
Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:
Reviewer checklist
📚 Documentation preview 📚: https://pytorch-lightning--20913.org.readthedocs.build/en/20913/