-
Notifications
You must be signed in to change notification settings - Fork 1.4k
feat: Make scaffolding Controller more generic #3408 #3416
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
feat: Make scaffolding Controller more generic #3408 #3416
Conversation
/bot run |
PR_Github #1604 [ run ] triggered by Bot |
PR_Github #1604 [ run ] completed with state |
424f777
to
aa2ad63
Compare
/bot run |
PR_Github #1674 [ run ] triggered by Bot |
PR_Github #1674 [ run ] completed with state |
aa2ad63
to
a6ba972
Compare
/bot run |
PR_Github #1686 [ run ] triggered by Bot |
PR_Github #1686 [ run ] completed with state |
a6ba972
to
fc07da9
Compare
bot run |
/bot run |
PR_Github #1939 [ run ] triggered by Bot |
PR_Github #1939 [ run ] completed with state |
Signed-off-by: fredw (generated by with_the_same_user script) <[email protected]>
fc07da9
to
ac41845
Compare
/bot reuse-pipeline |
PR_Github #2020 [ reuse-pipeline ] triggered by Bot |
PR_Github #2020 [ reuse-pipeline ] completed with state |
We need to do some optimizations to make the Controller more generic, which will help with better modularity.
These changes will not affect the existing API.
`
Then the "ScaffoldingLlm" will run the "process" of each controller with tasks and kwargs parallel. And the generation controller here can be someone which customized by user. In other words, users can pass any class of generation controller into majority vote controller without changing the code of majority vote controller.
We make the reward method in BestOfNController more generic as we observe that reward methods become various in recently paper and works.
We remove most of type check for Task as users may override them.