Skip to content

Installing Python tools can be challenging. Especially tools that rely on GPU processing through pytorch or tensorflow can be tricky. Also installing all your different software packages in one environment will quickly cause problems. Hence setting up separate Python environments is the best solution. A package manager helps to do so. Conda is such tool, and recen

pixi

You can read more here how to set pixi up and how to install tools such as Cellpose, Stardist with GPU support.
https://github.com/Leiden-Cell-Observatory/AI_tools_pixi

Setting up conda for Python

Installation

To be able to handle multiple Python software environments on a system use of conda is advised. Different software packages require different versions of Python and other dependencies. Conda allows to create isolated environments on your computer. It is recommended to always work within an conda environment and not in the base environment.

A detailed instruction how to setup conda can be found here.

https://biapol.github.io/blog/mara_lampert/getting_started_with_miniforge_and_python/readme.html#ref-miniforge-python