Sklearn-smith: an app to forge scikit-learn compatible estimator

Show the Community!
1

I recently created my first app with streamlit: a tool to generate boilerplate code for scikit-learn estimators.
It started as a CLI, and then developed into a streamlit app.

WebUI: https://sklearn-smithy.streamlit.app/
Repo: GitHub - FBruzzesi/sklearn-smithy: Toolkit to forge scikit-learn compatible estimators

Looking for honest and brutal feedback!

image
image1222×628 50.3 KB

1 Like
2

Great idea. Would be neat if it could generate some sample data to send quickly test it (filling in the optional parameters to match the sample dataset). Or even something preset so you could quickly test some sample code.

Maybe quick links to jupyterlabs in some cloud services : )

Nice idea to add the popup help all over.

1 Like
3

Hey @shawngiese , thanks for the feedback!
That looks significantly more complex and would require many more user inputs in my opinion. Maybe shaping the UI slightly different could serve it better.

Currently I am considering if integrating streamlit-code-editor to allow code editing on the web before downloading it. I am a bit uncertain of the library stability though

4

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.