Preparing Data for Machine Learning with Automunge v1.78

Have to admit it’s getting better

The Gap Band — I Don’t Believe You Want to Get Up and Dance

The following presentation is intended as an overview of features for the open source Automunge software for automated data wrangling to celebrate the rollout of our new version 1.78. There is also a companion Colaboratory notebook with code demonstrations available linked here, which allows a user to upload and process their own data if so desired. Cheers.

Colaboratory demonstration notebook

In short, this is what we do.

Structured Data (aka Tabular Data) = Tables of Cells

We make preparing data for machine learning very easy!

This is what we call “artificial learning.” Data is prepared for presentation to the machine learning algorithms in multiple formats.

I know what you’re thinking: “Why didn’t I think of that?”

Note that having more than 4 rows of data may help with accuracy.

Because sometimes we want to drive the car ourselves.

Pay attention to what your predictions are paying attention to.

Algorithms learn better with a balanced frequency of labels during training.

I mean this part not exactly rocket science but it’s certainly helpful.

Train your model today, process new data tomorrow. Easy!

Don’t tell anyone, this is a secret.

Follow us on twitter for announcements or for more check out automunge.com !

Colaboratory demonstration notebook

This presentation was intended as an overview of features for the open source Automunge software for automated data wrangling to celebrate the rollout of our new version 1.78. There is also a companion Colaboratory notebook with code demonstrations available linked here, which allows a user to upload and process their own data if so desired. Cheers.

Funkadelic — One Nation Under a Groove

*For further readings please check out my Table of Contents, Book Recommendations, and Music Recommendations. For more on Automunge:

• automunge.com
• github.com/Automunge
• twitter.com/automunge

* Patent Pending