Bison Lab

All analysis code related to the Bison Lab project at Earthshot Institute.

Code can run online using Binder

or locally using the setup below

Getting Started

Prerequisites

• An Anaconda Python distribution is required as this simplifies the installation. Either Miniconda or Conda are suitable: please follow the conda installation guide.

Setup

• In terminal/command line, navigate to your preferred software directory and clone this reposity to your local computer

# with SSH key
git clone git@github.com:earthshot-institute/bison-lab.git

# or with username/password
git clone https://github.com/earthshot-institute/bison-lab.git

Create a Conda (see prerequisites above if you don't have Conda installed) environment called bison-lab in which we can install all libraries/dependencies for this repository:

1. Change directory into the bison-lab project directory you cloned from GitHub with cd bison-lab.
2. Run the Conda command to create an environment based on the environment.yml configuration file in the bison-lab project directory.

conda env create -f environment.yml

3. Before running any code activate this conda environment:

conda activate bison-lab

4. (Optional) If you want to use the Jupyter Lab interface, first make sure the environment is activated (step 3), then install the widget to enable leaflet maps.

jupyter labextension install @jupyter-widgets/jupyterlab-manager jupyter-leaflet

🎉 You're ready to go! 🎉

The environment can be deactivated with:

conda deactivate

Code Quality

To run code quality tooling on every commit, we need to install the pre-commit Git hook:

pre-commit install

Now on every commit, the hooks configured in the .pre-commit-config.yaml will be executed.

Updates to Conda environment

If new Python libraries/packages are required, they should be added to the environment.yml file. These new packages are only installed with the creation of a new environment, however, you can update your existing Conda environment using this command:

conda env update --prefix ./bison-lab --file environment.yml --prune

Uninstall

To remove the bison-lab conda environment:

conda env remove -n bison-lab

Notebooks

Run Jupyter from the notebooks directory.

You have the option of using JupyterLab:

jupyter lab

or the "classic" Jupyter notebook:

jupyter notebook notebooks/