Editing issue with pickle def with lambda function #23869
Merged
+5
−1
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
sgugger
reviewed
May 30, 2023
src/transformers/optimization.py
Outdated
| @@ -32,6 +32,21 @@ | |||
| logger = logging.get_logger(__name__) | |||
|
|
|||
|
|
|||
| def get_constant_lambda(_=None): | |||
There was a problem hiding this comment.
Suggested change
| def get_constant_lambda(_=None): | |
| def _get_constant_lambda(current_step): |
src/transformers/optimization.py
Outdated
Comment on lines
36
to
45
| """ | ||
| Return 1, independent from args. | ||
|
|
||
| Args: | ||
| _ ( *optional*, defaults to None): | ||
| Placeholder to argument, used to consistency with args in LambdaLR | ||
|
|
||
| Return: | ||
| 1 : int - constant lambda for constant scheduler | ||
| """ |
There was a problem hiding this comment.
No need for a documentation for a private function.
src/transformers/optimization.py
Outdated
| @@ -46,7 +61,7 @@ def get_constant_schedule(optimizer: Optimizer, last_epoch: int = -1): | |||
| `torch.optim.lr_scheduler.LambdaLR` with the appropriate schedule. | |||
| """ | |||
|
|
|||
| return LambdaLR(optimizer, lambda _: 1, last_epoch=last_epoch) | |||
| return LambdaLR(optimizer, get_constant_lambda, last_epoch=last_epoch) | |||
There was a problem hiding this comment.
Suggested change
| return LambdaLR(optimizer, get_constant_lambda, last_epoch=last_epoch) | |
| return LambdaLR(optimizer, _get_constant_lambda, last_epoch=last_epoch) |
There was a problem hiding this comment.
done refactoring based on function name
added 2 commits
May 30, 2023 20:04
|
The documentation is not available anymore as the PR was closed or merged. |
sheonhan
pushed a commit
to sheonhan/transformers
that referenced
this pull request
Jun 1, 2023
* Editing issue with pickle def with lambda function * fix type * Made helper function private * delete tab --------- Co-authored-by: georgebredis <[email protected]>
gojiteji
pushed a commit
to gojiteji/transformers
that referenced
this pull request
Jun 5, 2023
* Editing issue with pickle def with lambda function * fix type * Made helper function private * delete tab --------- Co-authored-by: georgebredis <[email protected]>
novice03
pushed a commit
to novice03/transformers
that referenced
this pull request
Jun 23, 2023
* Editing issue with pickle def with lambda function * fix type * Made helper function private * delete tab --------- Co-authored-by: georgebredis <[email protected]>
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?
In this PR, I address the problem of pickling the constant LR scheduler, which fails during the process (potentially during multi-GPU training, as observed in my case) due to the presence of a lambda function within it.
Fixes #23865 (issue)