Add ModelPruning(prune_on_train_epoch_end) to choose when to apply pruning
#7704
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
carmocca
requested review from
Borda,
SeanNaren,
awaelchli,
justusschock,
kaushikb11,
tchaton and
williamFalcon
as code owners
carmocca
commented
May 25, 2021
| assert hasattr(pl_module.layer.mlp_3, "weight_orig") | ||
| if not on_train_epoch_end: | ||
| # these checks only work if pruning on `validation_epoch_end` | ||
| # because `on_save_checkpoint` is called before `on_train_epoch_end` |
Contributor
Author
There was a problem hiding this comment.
This is not true in current master, but will be after #7357
Codecov Report
@@ Coverage Diff @@
## master #7704 +/- ##
=======================================
- Coverage 93% 88% -5%
=======================================
Files 199 199
Lines 12960 12968 +8
=======================================
- Hits 11990 11379 -611
- Misses 970 1589 +619 |
awaelchli
approved these changes
May 25, 2021
Co-authored-by: Adrian Wälchli <[email protected]>
ananthsub
approved these changes
May 25, 2021
awaelchli
approved these changes
May 25, 2021
|
|
||
| def on_train_end(self, trainer: 'pl.Trainer', pl_module: LightningModule) -> None: | ||
| if self._make_pruning_permanent: | ||
| rank_zero_debug("`ModelPruning.on_train_end`. Pruning is made permanent for this checkpoint.") |
Contributor
There was a problem hiding this comment.
btw now you have a string here that has a period where there is no sentence and also a sentence but with a missing period. xD
Contributor
Author
There was a problem hiding this comment.
I guess I should've used : to separate haha
Not going to change it, not worth re-running CI just for it
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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?
See title. Same idea as:
https://github.com/PyTorchLightning/pytorch-lightning/blob/3e5d6e906a0165c59fc898e4c3da44a39771dbd0/pytorch_lightning/callbacks/early_stopping.py#L63-L64
Also, use regexes so we are not testing the seeding.
Last piece before #7357
Before submitting
PR review