[Feature Request]: Official Dockerfile / Docker-Compose about llm-finetuning-toolkit

[RohitSaha]

@anthonyprinaldi It'd be great if we can start with the fine-tuning piece. The current requirements.txt file contains all the packages one needs to finetune the models so would be nice if we can dockerize that bit first.

[RohitSaha]

@sammcj We finally have a dockerfile added to the repository.

@benjaminye can help answer any questions you may have.

Please let us know if you have more feedback to give. And we are always looking for external collaborators :)

[RohitSaha]

Thanks for the feedback! We will work on including this file soon.

[anthonyprinaldi]

@sammcj there are a few different tasks for the project (i.e., fine-tuning, evaluation, and inference). Which of these are of interest to you?

[sammcj]

@anthonyprinaldi personally I agree with Rohit, fine-tuning is something that currently isn't as easy for most folks to get into where as there's a lot of front ends for inference.