Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 18 Next »

Introduction: This workshop will discuss how the various Python related tools work together on the cluster and introduce a series of best practices for managing these environments.

Course Goals:

  • Understand where pip installs packages within a user’s home folder with respect to different versions of Python.

  • Understand how to use pip to install a package within a conda environment.

  • Demonstrate how to create a conda environment, with pip installs, that can be shared across a project.

  • Demonstrate creating a module file to load a conda/python environment.

  • Suggested Best Practices.

This is not a workshop on learning the Python language, but on using Python, Pip and Conda explicitly on the ARCC clusters (not on desktops). Users should be familiar with python, using pip, and creating Conda environments.

Notes:

  • The workshop modules work best in a sequential manner as a story introducing concepts and providing examples, but sections can be used separately to focus on a particular concept.

  • If you have existing Conda environments, or have installed your own version of Anaconda, then the examples within this workshop might provide different results.

  • You will need to modify username, project names, and folder locations to apply to yourself.


Sections

  1. Python Pip Installs on the Cluster: Understand where pip installs packages within a user’s home folder with respect to different versions of Python.

  2. Pip Install within a Conda Environment: Understand how Conda’s pip works with a User’s Python pip package Installs.

  3. Create a Shared TensorFlow Conda Environment: General process for creating and sharing a conda environment under a project.

  4. Create a Module File to Load Your Conda Environment: Demonstrate creating a module file to load a Conda environment.

  5. Extend Conda Environment to Jupyter Kernel: Demonstrate extending a Python related conda environment in to Juypter kernel.

  6. Conda and Pip Environments and Reproducibility: Introduce ideas and practices to assist in managing the reproducibility of environments created using Conda and Pip.

  7. Python, Conda and Pip: Suggested Best Practices: Suggest best practices and summarize workshop.


 

  • No labels