This seems like a pretty common use-case but I’m struggling to understand how everything works based on the documentation. I am new to the space so I’m hoping someone with a bit more experience can help me understand how I should think about implementing this.
I’d like to create my own model where:
Input - a prompt describing some behavior
Output - Code that accomplishes the behavior
I’d like to train the model in three ways:
- The coding language has API documentation, so I’d like to download this content and feed it into the model so it understands all of the API that exists and how it works.
- Provide open source Github repositories that are projects with high quality code for the model to understand expected output.
- Examples of prompt / code example pairs that serve as correct examples of the expected input and output.
Any assistance in how to approach this problem would be appreciated. Similar open source projects or video tutorials would be a plus. Thanks in advance!