Virtual Environments

EDS 220: Working with Environmental Data

What are environments?

A way to keep the packages and Python versions you use for different projects organized.

Why use them?

  • To not interfere with your computer’s pre-installed Python.

  • Packages usually depend on other packages to work properly, this is called a package dependency. Dependencies across different packages need to be carefully managed and may potentially be different across projects.

  • Reproducibility! Share your code and where ran it with others

Conda environments

Conda is an environment and package management system:

  • It can both create and administer the environments and install compatible versions of software and their dependencies.

  • Environments created with conda are usually called conda environments.

  • Can manage packages for any programming language (not just Python).

What are conda channels?

Conda channels are the locations where packages are stored.

By default, packages are downloaded and updated from the default channel. But there are others!

We can choose which conda channel to install a package from.

What about pip?

  • pip is a package management system for Python.

  • We can use it to install packages from the Python Package Index (PyPI). Only Python packages.

  • We can use pip inside a conda environment when a package is not available from a conda channel.