Machine Learning Web Application - predicting housing prices using the California Housing Prices dataset

I wanted to use Streamlit - a super-useful package which allows quick creation of working web apps for data using just Python - to showcase some of my recent experiments with Machine Learning.

A screenshot of the California Housing Prices Web Application

My goal was to build a working ML application which allows the user to adjust the input parameters and receive a prediction from the model. I used the California Housing Prices dataset out of a respect for ML tradition (and because it was easily available, the data were already cleaned and I had built models using it previously - those were all factors too!), but this approach is extendable to any ML model that makes a discrete prediction.

I was very impressed with Streamlit in terms of its flexibility and ease of use. I would definitely recommend giving it a look if you want to easily share the results of some of your Data Analysis work. My only quibble is that the look and feel are not customisable with HTML and CSS, as they would be with an app built in Flask, for example. But that is a deliberate choice, and more of a feature than a bug. It keeps the focus on the data and means our analyses can be shared more quickly, without the hassle (or, if you're like me, the fun!) of endlessly tweaking the HTML and CSS.

I deployed the app to the cloud using Heroku, which I recommend for quickly and (relatively) easily deploying web applications

You can check out the app itself, and take a look at the code on GitHub.