Validate that state-key is unique when using multiple callbacks of the same type #15634
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Outdated suggestions cannot be applied.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?
If you give two different callback instances with the same state key to the Trainer, they overwrite each other when saving a checkpoint. Only one of two states will actually be in the checkpoint. Hence, the trainer would fail to resume all states correctly (silently).
This PR validates this at the time of the trainer instantiation.
This change is required to continue with an improved version of #15606 to make checkpoint migration better.
Does your PR introduce any breaking changes? If yes, please list them.
If someone was affected by this but weren't noticing it, when they upgrade Lighting they will get a friendly error now. Apart from that change, there are no intended breaking behaviors.
Before submitting
PR review
Anyone in the community is free to review the PR once the tests have passed.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:
Did you have fun?
I made sure I had fun coding 🙃
cc @Borda @awaelchli