Python environment with useful tools

Some Python tools are required for using GeoClaw, such as numpy and matplotlib. There are many other Python tools that are useful for dealing with geophysical data and/or more advanced plotting. The file requirements_geo.txt in this directory contains a list of things you might want to pip install. (Feel free to edit it to add or delete things based on your needs.)

You are strongly encouraged to create a new Python environment before pip installing a lot of software, and some operating systems now require this. The instructions below might be useful.

Building a Python virtual environment¶

This procedure should work on a laptop (linux or MacOS) using the standard Python module venv.

mkdir ~/venv   # directory for virtual environments
python -m venv ~/venv/geo
source ~/venv/geo/bin/activate
python -m pip install --upgrade -r /path/to/requirements_geo.txt

Using the virtual environment¶

When open a new shell or terminal window, the command

source ~/venv/geo/bin/activate

should activate the environment. You might want to put this in your .bashrc file, for example. Then if you start python, ipython, or a Jupyter lab in this environment, the packates listed in requirements_geo.txt (and their dependencies) should be available to import.

Building a conda environment on DesignSafe¶

The following should work, but doesn’t seem to at the moment....

First launch JupyterHub from https://designsafe-ci.org/use-designsafe/tools-applications/analysis/jupyter/

Then start a terminal, and in this terminal:

conda config --add envs_dirs ~/MyData/Python_Envs  # directory for env
conda create --name geo  -y -c conda-forge pip python
conda activate geo 
python -m pip install --upgrade -r /path/to/requirements_geo.txt

pip install --src=$HOME/MyData/clawpack_src --user --no-build-isolation -e \
    git+https://github.com/clawpack/clawpack.git@v5.13.1#egg=clawpack