Adding PEFT compatibility #181
Open
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
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Outdated suggestions cannot be applied.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.
Please consider this a draft :)
I'd like to use LoRa for faster finetuning. Do you have advice on how I can best test whether the changes proposed in this PR are correct? Is there another dataset such as TheCauldron that is easily compatible with nanoVLM? Ideally I'd like to use the nanoVLM checkpoint that you've published on huggingface and that is already pretrained on TheCauldron and then see how LoRa based finetuning works for a new dataset.
Currently I'm using this LoRa setup to finetune a pretrained nanoVLM on a robotics dataset. It's learning but the results are inconclusive. It's been a day on a consumer GPU and the data is far out of distribution compared to the pretraining data, so I'm not expecting much. The loss is decreasing, but I just have no frame of reference. So this can't really work as a test whether my PEFT changes introduce subtle bugs.