PROCESS Examples

Examples of how to use PROCESS are provided in the form of Jupyter notebooks. These provide a convenient way of explaining PROCESS usage alongside runnable code cells, as well as displaying some types of output.

Installation

jupyter (which includes the notebook package) should already be installed in the process virtual environment when the cmake installation is performed (it is included in the requirements.txt).

If not, install with

pip install jupyter

for all Jupyter packages or

pip install notebook

for just the notebook package.

Notebooks in VS Code

The easiest way to run notebooks is in VS Code; this has the additional advantage of being able to debug notebooks. Simply open the .ipynb file in VS Code to view and run it. You may be required to select a notebook kernel on first run; be sure to select the virtual environment where PROCESS is installed (e.g. ~/process/.venv).

Notebooks via the Jupyter server

The original way of running Jupyter notebooks has been via a web browser and the Jupyter server. To launch the server, navigate to the examples directory and run:

python -m notebook

A web browser will open and the notebook can be run from there. If you're using WSL, you'll have to ctrl + click the link.

Notebooks in Docker

To access the notebooks when running inside of a Docker container the Jupyter server port must be forwarded out.

docker run -it -v ~/process/:/root/process -p 8888:8888 process_dev_image

The Jupyter server can then be run similarly to above:

python -m notebook --ip=0.0.0.0 --port=8888 --allow-root

Now the Jupyter server can be reached on http://127.0.0.1:8888/?token=<token> in a web browser, where <token> is a random password generated, and displayed, in the terminal when starting the server.

For some users, they will encounter a PDFInfoNotInstalledError inside of the Notebook (specifically noticed when running inside the dev image currently). This error can be fixed as such, until the image is updated:

apt-get update
apt-get install poppler-utils

Maintaining notebooks and scripts

Notebooks (.ipynb) and scripts (.py) with the same filenames are maintained in the examples directory. From within VS Code, a notebook can be exported as a Python script. The reason for this is so that notebooks are always available to run directly, whilst the corresponding Python scripts are much easier to track diffs in when reviewing changes (notebooks are ultimately JSON and so create difficult diffs). The scripts can also be tested by pytest to ensure the example notebooks keep working.