how to inference finetuned embedding?

[miniroll]

Guys, I’ve tried the code and look into it. I can’t see any inference related stuff. Looks interesting, but unless I can easily redeploy it in my infrastructure, it’s hard to be adopted.
Are you going to implement something like this? Or can you suggest any workaround?

[oYASo]

Hi! In the upcoming release, we'll be adding endpoints to display training metrics. Stay tuned for the update!