-
Notifications
You must be signed in to change notification settings - Fork 263
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.
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
Local tokenizer and processor for more consistent CI #16
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CI on this PR seems to work, so that's good!
Still seeing failures here even with local tokenizer/processor. I think something is getting wedged during test shutdown, kind of frustrating to debug since everything works fine locally 🙁 |
OK, so the issue here was ddp_utils_test.py, not infer_test.py. That said, I think this is worth keeping since it allows external PRs to access the Llama 3 tokenizer, which would otherwise fail due to not having the HF_TOKEN set. |
* saving local versions of tokenizer and processor for faster CI * cr * try again * and again * debug * verbose tests * Move files * Revert "verbose tests" This reverts commit bf63a43. * clean up * Reapply "verbose tests" This reverts commit aa09382. * more logging again * blarg * remove logs --------- Co-authored-by: juberti <[email protected]>
* breaking up ds_tool to make it more manageable/readable * move tts and chat client to their respective files
No description provided.