Added LMFormatEnforcer integration #65
Conversation
There was a problem hiding this comment.
Linking to the main repo that's not only a haystack integration is totally fine. If you like, you can make the repo link go to the file where the custom component is implemented.
A couple of notes:
You could add version: Haystack 2.0 into your frontmatter which will generate the 'Haytsack 2.0' badge on the website for you.
I will suggest some edits on certainl lines here too. Other than that, all looks good! I will maybe change the size of the image but we can see how this one looks once we merge it first.
|
Edited based on all of your comments. |
|
OK, Added a logo. I think we're ready |
|
@noamgat LGTM, I'm sorry I didn't spot this the first round, but if possible at the very top I think it would be useful to have a short description on why this component is useful and why to use it. Something like the intro paragraph on your own repo. I think it would be useful for people who don't know about why format enforcing matrers. Once that's done, I will merge this and it will go live 🚀 Thanks for the contribution! |
|
No problem, done! |
Following the discussion on the main haystack repo : deepset-ai/haystack#6222
This PR adds the LMFormatEnforcer integration.
Note: The pypi link is to the main repository, not a dedicated haystack integration one.
LMFormatEnforceris a relatively small library, and all of the integrations are lazily loaded and are not part of the project's dependencies. Therefore there is no "extra baggage" by just pointing to the library itself.