The aim of the repository is to efficiently implement the CURLoRA Research Paper within HuggingFace.
Contribution, new ideas, collaboration and any kind of help will be much appreciated.
When contributing to this repository, if possible but not required, please first let's discuss together the change you wish to make via issue, email, or any other method you prefer. I would like to learn from you.
There is still a lot to do to enhance the code and a lot to be implemented like:
- enabling CURLoRA with Quantization to perform QCURLoRA.
- Update the README.md with some explanation about the changes or fixes you made and/or added.
- Increase the version numbers in any examples files and the README.md to the new version that your Pull Request would represent.
Please also note that by contributing, you agree that your contributions will be licensed under MIT License.