The Yhat Blog

machine learning, data science, engineering

Brand New! Interactive Data Science Apps

by Elise Breda |

Today is a big day

We’re very excited to lift the curtain on a brand new project we've been working on on the side for the past few months.

How we got here

This June we published “Data Science in Practice,” a whitepaper about real-life examples of data science applications, from recommender systems to dynamic pricing and credit scoring. We talked about how companies all over the world develop and implement machine learning models into their products, including how our very own software, ScienceOps, drastically streamlines the process.

In the dozens of email, twitter, skype and phone conversations that followed, lots of you expressed a similar sentiment...“I get the concept, but please show me the code!”

Those are the folks we built this project for.

Interactive applications powered by ScienceOps

Instead of writing another whitepaper or even a blogpost to explain how data powered products works, we wrote five actual models and embedded them into web apps that you can explore for yourself.

We've included a variety of different use cases on the new page--including everything from predicting the contents of an image using neural networks to delivering a personalized beer suggestion using a product recommender. But regardless of the specifics, each of the apps calls R or Python models directly using ScienceOps.

Explore code powering web apps from image recognition to product recommenders

In addition to the actual javascript app that you can play around with, each of the pages also includes both the technical details e.g. POST/cURL requests, github links with full code repo...

...and a non-technical explanation, e.g. what you’re looking at, in plain english.

Goodbye farewell auf wiedersehen

You can find all these new web apps at or just navigate to Solutions > Explore from our homepage.

Feel free to send any questions or feedback my way ( Otherwise, we’ll leave the exploring up to you!

Our Products

Rodeo: a native Python editor built for doing data science on your desktop.

Download it now!

ScienceOps: deploy predictive models in production applications without IT.

Learn More

Yhat (pronounced Y-hat) provides data science solutions that let data scientists deploy and integrate predictive models into applications without IT or custom coding.