Using a virtual python environment to customize an environment
It is considered good practice to create and use a python virtual environment for python projects with dependencies not satisfied by the base environment. This prevents dependency conflicts between projects. Cerebras does not support conda, but python virtualenvs can be used instead.
python3 -m venv ~/testp3env source ~/testp3env/bin/activate pip install --upgrade pip # sample pip install pip install wget # to exit the virtual env: deactivate
To use this virtual env, e.g. in a script started with csrun_wse or csrun_cpu, or in a singularity shell:
Customizing the Cerebras singularity container
See this section of the Cerebras documentation: Adding Custom Packages To cbcore Container