...

# find a specific installation of cineca-ai module
module load spack
spack find <CINECA-AI PACKAGE>
e.g. spack find py-torch

# use the specific installation of cineca-ai module loading it via spack
spack loads <CINECA-AI PACKAGE>
e.g. spack loads py-torch

For checking the loaded packages and eventually unload them
spack find --loaded
spack unload --all

# use the specific installation of cineca-ai module to install additional packages via spack 
spack install <YOUR PACKAGE> ^<CINECA-AI PACKAGE>  
e.g. spack  install <YOUR PACKAGE> ^py-torch 

3. Notebook use in a Python environment

To open a notebook in a Python environment, it is neccessary to:

...

load the deeplrn profile: 

module load profile/deeplrn

load cineca-ai:

 module load cineca-ai

create your environment: 

python -m venv <name> --system-site-packages

activate your environment: 

source <name>/bin/activate

...

export PYTHONPATH=<path>/<name>/lib/python3.10/site-packages: $PYTHONPATH

...

create an SSH tunnel and start the notebook in your browser

 ssh -L 9999:localhost:9999 USER@login01-ext.leonardo.cineca.it -N

Step 1. Launch jupyter notebook

 On your REMOTE machine run the following command to open a jupyter notebook:

 jupyter notebook --port=9999 --no-browser

NB: once you have also completed step 2, open a browser on your local machine and copy the URL in the address bar.

Step 2. Open the SSH tunnel

 You are on the remote machine (or in dry run mode), so follow these instructions

 On your LOCAL machine open an SSH tunnel using the following command:

  ssh -L 9999:localhost:9999 <user>@login0<login_number>-ext.leonardo.cineca.it ssh -L 9999:localhost:9999 <compute_node> -N