Docs › Environments

Using Anaconda.

The cluster already has a shared Anaconda install anyone can use — you just need to initialize it once per account.

One-time setup

Copy the following into a file — call it conda.sh:

conda.sh 
export PATH=$PATH:/home/software/hess/anaconda3/bin:/home/software/hess/anaconda3/condabin  # access to anaconda
conda init bash    # for environment activation
conda config --set auto_activate_base false  # don't activate base by default

Then run it once:

Run it 
source conda.sh

Log out and back in. From then on, conda works normally without repeating this step.

Using environments

A few environments already exist on the shared install — see what's there:

List environments 
conda env list

Create your own environment tailored to what you need, then activate it:

Create and activate an environment 
conda create --name <environment name> python=<python version>
conda activate <environment name>
Install a package 
conda install <package>

For anything beyond this, the conda documentation covers the rest.

Working on H.E.S.S. analysis?

Don't hand-build a Paris Analysis environment from scratch — use the current pa10 alias described in Paris Analysis & H.E.S.S. access, which activates the right conda environment (and ROOT version) for you.