Orange — Data mining, Analysis and Visualisation tool for ML engineers and Data Scientists P.C.

What is Orange?

Orange is an open source, visual software programming package using which you can do data visualisation, data mining, machine learning and analysis of data. Its core is written in C++ with Python wrappers and is developed by the University of Ljubljana (Slovenia).

The default installation includes a number of machine learning, preprocessing and data visualisation algorithms in 5 widget sets (data, visualise, model, evaluate and unsupervised). Additional functionalities are available as add-ons (bioinformatics, data fusion and text-mining).

Why Orange (when we already have R and Python)?

Traditionally, R and Python have been the tools of choice for data scientists and machine learning engineers. However it takes considerable amount of time to learn and master these before practising data analysis techniques. ML engineers need to try out different models and visualisation techniques before arriving at accurate results. Orange offers to quickly portray visualisations via its intuitive GUI-style interface with widgets, menus, pointers, et al. It comes in as a handy tool for rapid data analysis and interactive data visualisation.


Prerequisites : Python 3

On mac, you can install Orange by downloading dmg file and running it

At the time of writing this article, the latest version was 3.27.1

Steps to upgrade Python from 2.x to 3.x

brew install python3

alias python=/usr/local/bin/python3

Fig 1: Execute dmg file (Mac)

Note: While trying to install Orange on windows, I faced numerous problems. Quick google search revealed that there are many who faced or are facing similar problems. After losing almost half a day trying to debug these issues, I decided to try it on Mac. It worked. Hence if you wish to just try this tool out, try it on Mac (or Linux).

Launch Orange

Fig 2: Home Screen of Orange when you launch the tool

There are 5 categories of widgets in Orange 3 -

  • Data (refer screenshot above)
  • Visualize
  • Model
  • Evaluate
  • Unsupervised

Widgets in each categories are given in Fig 3 below

Fig 3: Widgets in each category

Thats all for this tutorial.

In the next part, we will try out a sample prediction tutorial using Orange.

Engineer and Water Color Artist @toashishagarwal