[Question]: Similarity score of chroma: ...

interesting, I guess this slipped by. Should be fixed in a PR

We just want the score to be 0-1 where 1 is the most similar and 0 the least

Sure that's was I imagined. Can I open a pull request linking this issue?

Maybe the correct formula should be e^(-x)

Definitely open a PR!

Hmm actually I might fix this right now -- want to get this in before cutting a release lol

No problem Logan, thanks 🙂