-
Notifications
You must be signed in to change notification settings - Fork 27.4k
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
examples/rag: test coverage, tiny model #7715
Comments
sign me up, Sam |
examples/rag/finetune.py is not that stable. Seems like it depends on the pytorch_lightning version also. It would be nice if we can test it properly. |
I think this is still waiting for: #8284 to complete the missing info and perhaps some tests were added since then? |
This issue has been stale for 1 month. |
As the required to implement this info was never provided and I since then moved to work on other things I removed self-assignment to this ticket... |
Disclaimer: I don't know this code very well, this may be much harder than it seems.
Blocking PR: #7713
[
examples/rag/finetune.py
,examples/rag/finetune.sh
,eval_rag.py
] do not seem to be tested at all.It would be good to have a
test_finetune.py
likeexamples/seq2seq
that tested these.cc @stas00 if interested, rag is a cool new retrieval model https://arxiv.org/pdf/2005.11401.pdf
The text was updated successfully, but these errors were encountered: