search

KE7 / ScenicNL

Other
5 stars 1 forks source link

Fine-tuning #2

Open KE7 opened 1 year ago

KE7 commented 1 year ago

This is very high level as much more scaffolding needs to be implemented but thing of this as an "epic" from scrum methodology.

We would like to call a fine-tuned GPT3.5 model. The model should be fine-tuned on at least the Scenic codebase and documentation. Maybe even the CARLA documentation as well but CARLA is TBD. We can open a new issue for that.

https://platform.openai.com/docs/guides/fine-tuning

- Fine-tuning on the scenic codebase: 
- Function calling: https://platform.openai.com/docs/guides/gpt/function-calling
KE7 commented 1 year ago

notebook has been started and is in PR. pending GPUs