Add instruct tuning support to LoRA training #1211
Open
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
Fixes ml-explore#484 Add support for instruct tuning with input/output pairs and alternative loss functions. * **llms/mlx_lm/lora.py** - Add `CompletionsDataset` class to support input/output pairs. - Modify `Dataset` class to handle different dataset types. - Update `main` function to include new dataset type. * **llms/mlx_lm/tuner/trainer.py** - Modify `default_loss` function to support alternative loss functions. - Add new `instruct_loss` function for instruct tuning. * **llms/mlx_lm/LORA.md** - Add instructions for instruct tuning with input/output pairs. - Update documentation to include alternative loss functions. * **llms/tests/test_datasets.py** - Add tests for `CompletionsDataset` and `create_dataset` functions. * **llms/tests/test_trainer.py** - Add tests for `default_loss` and `instruct_loss` functions. --- For more details, open the [Copilot Workspace session](https://copilot-workspace.githubnext.com/ml-explore/mlx-examples/issues/484?shareId=XXXX-XXXX-XXXX-XXXX).
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Outdated suggestions cannot be applied.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.
Fixes #484
Add support for instruct tuning with input/output pairs and alternative loss functions.
llms/mlx_lm/lora.py
CompletionsDatasetclass to support input/output pairs.Datasetclass to handle different dataset types.mainfunction to include new dataset type.llms/mlx_lm/tuner/trainer.py
default_lossfunction to support alternative loss functions.instruct_lossfunction for instruct tuning.llms/mlx_lm/LORA.md
llms/tests/test_datasets.py
CompletionsDatasetandcreate_datasetfunctions.llms/tests/test_trainer.py
default_lossandinstruct_lossfunctions.For more details, open the Copilot Workspace session.