Updated Fabric trainer example to not call self.trainer.model during validation
#19993
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
Before, using the trainer would result in: AttributeError: Your LightningModule code tried to access `self.trainer.model` but this attribute is not available when using Fabric with a LightningModule. Which would occur during validation when 'on_validation_model_eval' and 'on_validation_model_train' were called, as they do not work with fabric. This change allows the model to be switched between train and eval mode (which the original behavior intends) without relying on calling 'self.trainer.model'. A check is added for overridden methods to not change a user's code.
liambsmith
requested review from
lantiga,
Borda,
tchaton,
awaelchli and
justusschock
as code owners
for more information, see https://pre-commit.ci
|
Added a second commit to fix a minor typo. |
awaelchli
reviewed
Jun 20, 2024
…al() Change to not overcomplicate the fix for fabric not working with the default "on_validation_model_eval" and "on_validation_model_train".
for more information, see https://pre-commit.ci
awaelchli
approved these changes
Jun 21, 2024
awaelchli
changed the title
self.trainer.model during validation
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Outdated suggestions cannot be applied.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.
Before, using the trainer would result in:
AttributeError: Your LightningModule code tried to access
self.trainer.modelbut this attribute is not available when using Fabric with a LightningModule.Which would occur during validation when 'on_validation_model_eval' and 'on_validation_model_train' were called, as they do not work with fabric. This change allows the model to be switched between train and eval mode (which the original behavior intends) without relying on calling 'self.trainer.model'. A check is added for overridden methods to not change a user's code.
Fixes #19992
📚 Documentation preview 📚: https://pytorch-lightning--19993.org.readthedocs.build/en/19993/