Meet Playground - my new Streamlit App

Hello everyone,

I made a YT video on how to build and deploy a streamlit application that allows you to interactively play with machine learning models directly from the browser.

The app is called playground: you can use it to test different models with different hyper-parameters on a set of non-linear classification problems.

Playground is meant for educational purposes to get familiar with models behaviour, notions of over/underfitting, drifts and to build an intuition behind each hyper-parameters. It doesn’t replace textbooks, it’s rather here to complete your knowledge. So take it as it is !:slightly_smiling_face:

Want to see a demo? check out the first minutes of this video: https://youtu.be/htKmCWrFYr8

Here’s the code: https://github.com/ahmedbesbes/playground

Want to try the app: it’s right here (desktop is better):

https://share.streamlit.io/ahmedbe…/playground/main/app.py

Feedbacks are welcome :slight_smile:

demo

10 Likes

Hey @ahmedbesbes,

WOW, THIS IS SO COOL!! :astonished: :astonished: :astonished:

I really love how you can so clearly see the differences between using different models! and that you can see the length of time it took to train the model, accuracy and F1 score!

This could be great for people just starting out in ML, wonderful job! :star_struck:

Happy Streamlit-ing!
Marisa

1 Like

@Marisa_Smith thank you so much! I hope to make it one day to the streamlit gallery :star_struck:

3 Likes

Well well well, let me just see what I can do! :wink:

1 Like

Hey @ahmedbesbes,

Look what I see here! :eyes:

2 Likes

@Marisa_Smith :star_struck::star_struck: amazing! many thanks!

1 Like