/psana_jupyter_kernel

Primary LanguageShell

PSANA Jupyter Kernel

This repository contains instructions on how to build, install and use a PSANA Conda environment as a python Jupyter Lab kernel.

NERSC usage

To activate the pre-built PSANA conda environment at NERSC to use it at https://jupyter.nersc.gov, just follow these instructions:

git clone https://github.com/llanaproject/psana_jupyter_kernel
cd psana_jupyter_kernel
module load python/3.7-anaconda-2019.07
pip install ipykernel
python -m ipykernel install --user --name psana2_py37 --display-name psana2_py37
cp kernel.json $HOME/.local/share/jupyter/kernels/psana2_py37

Now, you should be able to select PSANA from your list of kernels:

alt text

Building from scratch

In the sections below we show how to build this conda environment from scratch. There are three files to make it:

  1. build_from_scratch.sh file that clones and build psana2 to work with NERSC infrastructure.
  2. A Conda yaml file called env_create.yaml to create a Conda environment for PSANA.
  3. A kernel.json example file to use the Conda environment as a Jupyter Lab kernel.

Conda environment

git clone https://github.com/llanaproject/psana_jupyter_kernel
cd psana_jupyter_kernel
export DIST=$PWD
chmod +x build_from_scratch.sh
./build_from_scratch.sh

Installing the Conda environment as a Jupyter Lab kernel

After building the environment, the kernel has to be activated in our NERSC account:

source $DIST/env.sh
conda activate $DIST/psana2_py37
python -m ipykernel install --user --name psana2_py37 --display-name psana2_py37

This will create a directory on $HOME/.local/share/jupyter/kernels/psana2_py37 with a kernel.json file. Change its content to

{
 "argv": [
   "/path/to/start_kernel.sh",
  "-f",
  "{connection_file}"
 ],
 "display_name": "Psana_python37",
 "language": "python"
}

Or simply copy the file provided in this repository, and change it accordingly to your needs:

cp $DIST/kernel.json $HOME/.local/share/jupyter/kernels/psana2_py37

The last step consists in changing the variable DIST in start_kernel.sh#L3 to its correct path.

At this point, you might be able to go to https://jupyter.nersc.gov and Psana_python37 should be shown as a kernel in your Jupyter Lab.