Ted Sanders

Adding to my to-do list. Thanks!

It's been updated, thanks to a community PR. Let me know if you run into any issues.

Thanks! Seems unnecessary to have folks download the dataset when a jsonl is included with the repo. What do you think about including the code for reproducibility but commenting it...

Inherent randomness sounds like the most likely culprit. Those sound like small differences. Nothing is intended to change over time.

I tried to reproduce but it works fine for me. I see that your sklearn package says that it's deprecated and to use scikit-learn instead. Suspect that might be your...

Try installing scikit-learn and see if that fixes things for you. Thanks for including the helpful screenshots!

Will add soon. Thanks!

Added here (https://github.com/openai/openai-cookbook/blob/main/examples/data/fine_food_reviews_with_embeddings_1k.csv) in PR #66 and updated examples to use the precomputed embeddings. Let me know if you think anything more will be helpful. Will close now.

Thanks! Will fix.

Looks like `n_init='auto'` requires scikit-learn 1.2.0+. I'll omit it so that it just uses the default for whichever version of scikit-learn folks are using.