Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Alternative backends #3

Open
benwr opened this issue Jan 6, 2023 · 3 comments
Open

Alternative backends #3

benwr opened this issue Jan 6, 2023 · 3 comments

Comments

@benwr
Copy link
Owner

benwr commented Jan 6, 2023 •
edited
Loading

The current OpenAI Codex backend might not be ideal for everyone. It's working great for me, so I'm unlikely to find the time to add backends myself, but if you can find an alternative API to use, I'd be happy to accept PRs that add support.
@benwr
Copy link
Owner Author

benwr commented Jan 11, 2023 •
edited
Loading

I've looked around a bit for models that support this, and unfortunately I think it's pretty hard to find any.

The easiest option might actually be to train a language model on diffs with the message placed at the end of the sequence. That way we could at least re-use basically any popular architecture.

@benwr
Copy link
Owner Author

benwr commented Jan 11, 2023

It might be good to generate a diff-focused dataset anyway; I don't know how much of Codex's training set is diffs, but surely it's not a primary focus.

@benwr
Copy link
Owner Author

benwr commented Mar 26, 2023

This is especially urgent now that OpenAI (I think?) killed code-davinci-002

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@benwr