finetune_lora upgrades #2086
Open
finetune_lora upgrades #2086
+866
−15
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
for more information, see https://pre-commit.ci
…; fixed test_cli and other errors
for more information, see https://pre-commit.ci
…s/litgpt into finetune_lora_upgrade divergent
|
UPDATE: tested it in a 4xA100 environment, works well for multi-gpu setup |
Borda
approved these changes
Jul 8, 2025
Essentially it's the exact lora implementation before this PR. The changes are quite breaking so my rationale is it may be apt to still give users the option to run the previous version of lora.py, hence the name |
KaelanDt
requested changes
Jul 9, 2025
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Outdated suggestions cannot be applied.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.
Integrated some improvements to finetune_lora based on @lantiga's reference implementation.
Most of the optimizations are there, but I did not include the model registry feature because litgpt does not currently support litmodels and adding litmodels support is a PR in itself IMO.
Still WIP as I need to run some experiments (and potentially having to modify some test cases)