Hi,

I am currently interested in using my own dataset to fine-tune chroma for a specific use case. However, I have noticed that the training code is not available in the repository. I believe that having access to the training scripts would greatly enhance the project's utility and allow the community to contribute more effectively.

Would it be possible for you to consider releasing the training code or providing guidance on how to train the model with custom data? I understand that this might require additional time and effort, but I believe it would be a valuable addition to the project.

Thank you for considering my request.