Yoav Katz

> Thanks for the PR! > > In addition to the comments I made, > > * you may also want to add unitxt dependancies to `pyproject.toml` that users can...

> > Thanks for the PR! > > In addition to the comments I made, > > > > * you may also want to add unitxt dependancies to `pyproject.toml`...

Hi @lintangsutawika @haileyschoelkopf - I updated the documentation: https://github.com/EleutherAI/lm-evaluation-harness/pull/1615/files#diff-556af2425890013f9dd8057bc56bb15013d24813bb65981374bf542438156e68 and the pyproject.toml file.

Hi @pawelknes . Thanks for the PR. It looks very good. Since the issue started, we saw the need for this expand beyond metrics. For example, a person added a...

> > Hi @pawelknes . Thanks for the PR. It looks very good. > > Since the issue started, we saw the need for this expand beyond metrics. For example,...

Pawel - thanks for this PR. It's an important one, especially since we have LLM as judges. I've made some additions and changes to the PR - I'll be glad...

@pawelknes - I think the next steps are: 1. Remove allow_passing_data_to_remote_api from settings. 2. In class InferenceEngine(abc.ABC, Artifact): Make sure infer() checks data classification of the inferred instance (best to...

@OfirArviv - Overall loogs good to me, but tests seem to fail. 1. AssertionError: Error while trying to run IbmGenAiInferenceEngine. Please set the environment param 'GENAI_KEY'. We should not set...

Hi. I think primary what's left is to update the test value, and move the import to where it is used. Is this right?

> > Hi. I think primary what's left is to update the test value, and move the import to where it is used. Is this right? > > We have...