DCC workshop: Package and Environment Management for Reproducible Research Software

The goal of this workshop is:

• To understand the importance of environment management in ensuring reproducibility.
• To learn how to use popular tools for managing packages and environments.
• To gain hands-on experience in setting up and managing environments.

Program Outline:

1. Introduction to Package and Environment Management (15 minutes)

• Why Environment Management Matters: Discuss the importance of reproducibility in research and how proper environment management plays a crucial role.
• Challenges in Managing Dependencies: Overview of common issues like dependency conflicts, version control, and platform differences.
• Overview of Tools: Brief introduction to the tools covered in the workshop: Conda, Virtualenv, Pipenv, and Docker.

2. Managing Python Environments with Conda 

• Introduction to Conda: What is Conda, and why use it?
• Creating and Managing Environments:
  • Creating a new environment.
  • Installing packages within an environment.
  • Activating and deactivating environments.
  • Exporting and sharing environments (environment.yml).
• Hands-on Activity: Participants create and manage a Conda environment for a small Python project.

3. Virtualenv and Pip for Environment Management 

• Introduction to Virtualenv: What it is and how it differs from Conda.
• Using Virtualenv:
  • Setting up a Virtualenv environment.
  • Installing packages using Pip.
  • Freezing and exporting requirements (requirements.txt).
• Introduction to Pipenv: Overview of how Pipenv integrates Pip and Virtualenv.
• Hands-on Activity: Participants set up an environment using Virtualenv or Pipenv.

4. Containerization with Docker (30 minutes)

• Introduction to Docker: What is Docker, and why is it useful for environment management?
• Creating and Using Docker Containers:
  • Basics of Dockerfiles and containerization.
  • Building a Docker image for a Python project.
  • Running and managing Docker containers.
  • Sharing Docker images and containers.
• Hands-on Activity: Participants create a simple Docker container for a project.

5. Best Practices and Tools Comparison 

• Best Practices: Tips for choosing the right tool for your project and ensuring reproducibility.
• Tools Comparison: Quick comparison of Conda, Virtualenv, Pipenv, and Docker, including when to use each.

Preparation and Requirements:

• Participants should bring: A laptop with Python installed.
• Pre-installed Software:
  • Conda (via Anaconda or Miniconda)
  • Docker
  • Pipenv and Virtualenv