Component/generate embeddings #520
Conversation
There was a problem hiding this comment.
Thanks @tillwenke. I've left a few comments.
Would you up for adding a simple test case to your component?
Based on your component code I can not see why your pipeline is failing. I think we have to look on your whole pipeline to see what is going wrong.
done |
| self.model_provider = model_provider | ||
| self.model = model | ||
|
|
||
| to_env_vars(api_keys) |
There was a problem hiding this comment.
Could you extend the readme on how to use the different model providers? E.g. in how the api key environment variable should be named. I can assume that this differs from provider to provider.
| msg = f"Unknown provider {model_provider}" | ||
| raise ValueError(msg) | ||
|
|
||
| @retry() # make sure to keep trying even when api call limit is reached |
There was a problem hiding this comment.
Does this keep retrying infinitely? Would be nice if this can be somehow be modified to log when a connection limit is reached.
components/generate_embeddings/tests/generate_embeddings_test.py
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Thanks @tillwenke! There are still some open comments but I will merge this PR since we want to use this component for the chapter conference, I will make some changes to it and tag you in later PRs
Besides: