Short Course on Fastscape (ISTerre, Grenoble, October 19 to 21, 2021)

This repository contains a collection of Jupyter notebooks for Jean Braun's short course on Fastscape given in October 2021 at ISTerre, Universite Grenoble-Alpes.

More information concerning FastScape on this website: https://fastscape.org

• How to run the notebooks?
  • Run in the cloud (Binder)
  • Install and run locally (Docker)
  • Install and run locally (Conda)
• How to contribute?

How to run the notebooks?

Run in the cloud (Binder)

You can run the notebooks in your browser without installing anything thanks to binder. Just follow the link below or click on the "launch binder" badge above and it will launch remotely a new notebook server for you:

• Run on binder

This service is for demo purpose only, do not rely on it for doing more serious work.

Install and run locally (Docker)

Docker images are built automatically for this repository. Those images provide the whole computing environment, pre-installed and pre-configured for running the notebooks. The only requirement is to have Docker installed on your machine. It is available on all platforms Linux/Windows/Mac and it can be installed from the Docker website or using one of your platform's package managers.

Run the command below in a terminal to first pull the latest image (note: the Docker application must be running, you might need to launch it first):

$ docker pull fastscape/isterre-short-course:latest

Then run the command below to start the Jupyterlab application from the Docker container. Replace short-course-fastscape by any other name you want to give to your local container (optional). Also Replace /path/to/local-notebook-folder by the full path to the directory on your machine where you want to create/copy, edit and permanently store notebooks for this training course.

$ docker run \
    -it \
    --name short-course-fastscape \
    -p 8888:8888 \
    -v /path/to/local-notebook-folder:/home/jovyan/my-local-folder \
    fastscape/isterre-short-course \
    jupyter lab --ip 0.0.0.0

You can then enter in your browser the url and token provided to start using the Jupyterlab application.

You may want to copy the notebooks folder in your local working folder mounted in the docker container as my-local-folder. Open a terminal in Jupyterlab and run the following command:

$ cp -R notebooks my-local-folder/

When you are done you can stop and remove the container:

$ docker stop short-course-fastscape
$ docker rm short-course-fastscape

Troubleshooting

The url I entered in my browser doesn't point to Jupyterlab

You may already have another application running on localhost using the port 8888. Try another port when running the docker run command above, e.g., using -p 8889:8888. You also need to change the port in the entered url accordingly (e.g., localhost:8889).

The url I entered in my browser gives a page asking for a token

Copy and paste the token given in the url. If the token is invalid, you may have another Jupyterlab application already running on your machine. Try using another port as described above.

Check Docker's documentation for additional command line help and options.

Install and run locally (Conda)

Assuming that you have git and conda installed, you can install all the packages required to run the notebooks in a new conda environment using the following commands:

$ git clone https://github.com/fastscape-lem/isterre-short-course
$ cd isterre-short-course
$ conda env create -f environment.yml
$ conda activate isterre-short-course

Note: you could use mamba instead of conda. mamba is a faster alternative to conda.

Note: If you have a new Mac with an M1 processor. FastScape has not been ported to this architecture yet. So you will need to install the intel version. This means that you need to install the intel version of all other necessary packages in your environment. For this, you need to change the command:

$ conda env create -f environment.yml

to:

$ CONDA_SUBDIR=osx-64 conda env create -f environment.yml

Finally run the command below to start the Jupyterlab application. It should open a new tab in your browser.

$ jupyter lab

How to contribute?

Your contribution is welcome! Your can do so by reporting issues, suggesting new notebook examples or improvements to the current examples.

A few extra steps are required to prepare your contributions. You can first update the conda environment using the following command:

$ conda env update -n isterre-short-course --file environment-dev.yml 

This installs a few additional packages like pre-commit, which is used to ensure that all notebook cell outputs are cleared before adding or updating notebooks in this git repository. Run the command below to enable pre-commit (you only need to do this once):

$ pre-commit install

The script below is useful if you want to ensure that all notebooks are running without error:

$ python execute_all_notebooks.py

This script (as well as a script to build the Docker image) is run each time you open or update a pull-request on GitHub.