Python For Data Science

A Python tutorial for (beginning) data scientists with a non-computer-science background.

Introduction

Let's take a look at the following snippet of Pandas code:

If this makes you curious about:

• how loc[] works
• how it processes these expressions inside the square brackets
• how these mystical lambda expressions work
• what part of your input data is pointed to by these df references,

then this course is right for you!

Instead of offering a comprehensive guided tour through the Pandas or Scikit-Learn APIs, this tutorial aims at empowering you to understand how these tools work by explaining the fundamental programming constructs from which these tools are built. This will allow you to:

• effectively learn parts of the APIs that are relevant to you without supervision
• write correct code with fewer bugs by knowing what you're doing
• write readable and testable code, so your colleagues know what you're doing
• create your own tools

Prerequisites (Skills, Knowledge)

To fully enjoy this tutorial, you are expected to:

• Be able to clone this repository using any git client
• Have some programming experience
• Have some conceptual understanding of basic data analysis such as filtering/selection, grouping by, aggregation, transformation

While the tutorial starts with an introduction into the most basic Python concepts, the pace will be too high for people with no coding experience whatsoever.

Prerequisites (Technical)

To run this tutorial, you need at least:

• Any computing environment with Python >= 3.7 installed and available
• Recent Pandas (>= 1.0) and Numpy packages available globally or in a virtual environment
• Jupyter Lab (or simple Notebook) installed globally or in a virtual environment
• Any git client that allows you to clone this repository

In case of any trouble meeting these prerequisites, this article at Real Python can be helpful. I suggest using a pyenv+pipenv - based environment or a conda - based environment and not creating a mix of these approaches.

Use this tutorial by cloning this repository into a directory of choice and running jupyter lab or jupyter notebook from that directory.