Getting started¶

This mini-tutorial will help you set up a working Python environment on a Linux system. This will allow you to work with other peoples Python tools and start writing your own scripts.

Some concepts¶

First, let’s go over a few important concepts.

What is Python?¶

Python is a widely used general-purpose programming language, known for its emphasis on code readability.

How do I get Python?¶

Python comes preinstalled on Mac OS, most Linux distributions and is easily installed on Windows.

Warning

The system often depends on the preinstalled Python installation for various operations. Modifying the system installation is not recommended.

Instead of using the preinstalled version of Python, it is recommended to work using what are called virtual environments.

Setting up and managing virtual environments, including package management, is the focus of this mini-tutorial.

What is a Python package?¶

When you write code, you may want to use some parts other people have written, rather than rewriting everything from scratch. If you reference and external code in your code, we say that your code depends on that external code.

In Python, people often package up their code and make it available on the Python Package Index for others to download and use in their programs.

Installing miniconda3¶

In this tutorial, we will use a lightweight installation of conda called Miniconda3 to manage our software environments.

Note

conda is not Python specific, it can be used as a general software environment solution. This is useful if you use software which has many specific dependencies and requires activation prior to usage.

Creating a new conda environment¶

The following command will create a new conda environment called py38 with Python version 3.8.

conda create --name py38 python=3.8


This environment can be activated with conda activate py38 and deactivated with conda deactivate.

Installing packages¶

Once your environment is active, any packages you install will be installed into this environment, isolated from your system Python installation and other conda environments.

Packages can be installed with either conda install <package-name> or pip install <package-name>.