fix llm as judge warnings
Now loads openai lazily. the dep is no linger required when using lighteval
You could add a warning/error when loading metrics at the task level, to check there the deps if needed.
You could add a warning/error when loading metrics at the task level, to check there the deps if needed.
you mean like I did in task.py to check that the token is set ?
And for both of these, I mean using the same checks as for all the other optional deps :) (using new functions defined in utils)
@NathanHB I added the import testing I was thinking about - I'll let you review and if it works for you you can merge