QOL improvements and doc updates #1318
Merged
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
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
stevhliu
approved these changes
Jan 3, 2024
There was a problem hiding this comment.
Thanks for the updates! I would also add a short Utilities section in the LoRA API reference like this:
## Utilities
[[autodoc]] utils.get_linear_layer_names
BenjaminBossan
requested changes
Jan 10, 2024
src/peft/utils/other.py
Outdated
| @@ -498,3 +498,23 @@ def id_tensor_storage(tensor: torch.Tensor) -> Tuple[torch.device, int, int]: | |||
| unique_id = storage_ptr(tensor) | |||
|
|
|||
| return tensor.device, unique_id, storage_size(tensor) | |||
|
|
|||
|
|
|||
| def cast_non_trainable_to_dtype(model, dtype): | |||
There was a problem hiding this comment.
Just noticing that we may want to use a different name, since the trainable parameters will also be cast. The name could give the impression that trainable parameters are left untouched.
There was a problem hiding this comment.
Updated the PR to use cast_mixed_precision_params, let me know if you have better naming suggestions.
Co-authored-by: Benjamin Bossan <BenjaminBossan@users.noreply.github.com>
…ce/peft into smangrul/doc-updates-1
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Outdated suggestions cannot be applied.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?
TaskTypecast_non_trainable_to_dtypeto cast all non-trainable parameters of the model to the givendtype.This is meant to reduce the GPU memory usage when using PEFT methods by using half-precision dtype for non-trainable parameters.