ClusterLens: an open-source "explain my clusters" toolkit + docs built with Streamlit

Show the Community!
, ,
1

Hey everyone,

I’ve been working on an open-source library called ClusterLens, and I just published a Streamlit documentation app for it. Wanted to share it here and get feedback from the community:

What is ClusterLens?

ClusterLens is an interpretability engine for clustered / segmented data.

It assumes you already have clusters (k-means, GMM, HDBSCAN, rule-based segments, customer personas, risk bands, etc.), and then focuses on the harder questions:

  • What actually drives each cluster?

  • How is Cluster 1 really different from Cluster 3?

  • Which features make Cluster A “high value”, “at risk”, or “high cost”?

  • How do I turn all this into plain-English narratives for stakeholders?

I’d love your feedback:

If you:

  • Work with customer segments / user personas / risk bands, or

  • Are tired of re-writing the same cluster-exploration in every notebook,

Then, I’d really love your thoughts:

  • Is the API intuitive?

  • Are the visuals and narratives helpful for business stakeholders?

  • What’s missing for your use case?

Issues, ideas, and PRs are very welcome on GitHub: https://github.com/akthammomani/ClusterLens

And if you just want to chat or connect: https://www.linkedin.com/in/akthammomani/

Thanks for reading, and happy clustering!

1 Like