Add assertions to TPU accelerator for device availability #11799
Conversation
This file contains 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
ananthsub
requested review from
awaelchli,
justusschock,
kaushikb11,
SeanNaren,
tchaton,
williamFalcon,
Borda,
carmocca and
rohitgr7
as code owners
four4fish
approved these changes
Feb 8, 2022
carmocca
approved these changes
Feb 8, 2022
ananthsub
force-pushed
the
feat/tpu-validation
branch
from
136d174 to
88e65e8
Compare
rohitgr7
approved these changes
Feb 8, 2022
| @@ -98,6 +98,9 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). | |||
| - Added a `_Stateful` support for `LightningDataModule` ([#11637](https://github.com/PyTorchLightning/pytorch-lightning/pull/11637)) | |||
|
|
|||
|
|
|||
| - Added checks to `TPUAccelerator.setup_environment` to assert device availability ([#11799](https://github.com/PyTorchLightning/pytorch-lightning/pull/11799)) | |||
There was a problem hiding this comment.
Suggested change
| - Added checks to `TPUAccelerator.setup_environment` to assert device availability ([#11799](https://github.com/PyTorchLightning/pytorch-lightning/pull/11799)) | |
| - Added checks to `TPUAccelerator` to assert TPU device availability ([#11799](https://github.com/PyTorchLightning/pytorch-lightning/pull/11799)) |
| """ | ||
| Raises: | ||
| MisconfigurationException: | ||
| If the TPU device is not available. |
There was a problem hiding this comment.
Suggested change
| If the TPU device is not available. | |
| If the `torch_xla` is not installed. |
| MisconfigurationException: | ||
| If the TPU device is not available. | ||
| """ | ||
| if not _XLA_AVAILABLE: |
There was a problem hiding this comment.
should we add another one for TPU_AVAILABLE?
#3104
tchaton
approved these changes
Feb 9, 2022
| If the TPU device is not available. | ||
| """ | ||
| if not _XLA_AVAILABLE: | ||
| raise MisconfigurationException("The TPU Accelerator requires torch_xla and a TPU device to run.") |
There was a problem hiding this comment.
Suggested change
| raise MisconfigurationException("The TPU Accelerator requires torch_xla and a TPU device to run.") | |
| raise MisconfigurationException("The TPU Accelerator requires the `torch_xla` library and `TPU devices` to run.") |
11 tasks
|
Closing out in favor of the approach in #11797 |
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?
Add assertions to TPU accelerator for device availability
This aims to simplify the accelerator connector logic and rewrite effort in #11448
This would move the assertion logic from Trainer constructor checks to runtime at trainer.fit/validate/test/predict calls
Related PR: #11797
Does your PR introduce any breaking changes? If yes, please list them.
Before submitting
PR review
Anyone in the community is welcome to review the PR.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:
Did you have fun?
Make sure you had fun coding 🙃