Installation

RADWave is a pure Python package and can be installed via pip as well as Docker.

Dependencies

You will need a working Python 3+ (though it will work with Python 2.7) and the following packages are required: numpy, scipy, pandas, scikit-image, seaborn, geopy, cartopy, netCDF4, shapely, pymannkendall.

The complete list of Dependencies is available in the requirements.txt file and looks like:

numpy>=1.15.0
pytest
six>=1.11.0
setuptools>=38.4.0
pandas>=0.25
seaborn>=0.9
matplotlib>=3.0
geopy>=1.20
cartopy>=0.17
scipy>=1.3
netCDF4>=1.5.1
shapely>=1.6.4
scikit-image>=0.15
pymannkendall>=0

Installing using pip

You can install RADWave package using the latest stable release available via pip!

pip install RADWave

If you need to install it for different python versions:

$ python2 -m pip install radwave
$ python3 -m pip install radwave

To install the development version: Clone the repository using

git clone https://github.com/pyReef-model/RADWave.git

Navigate into the RADWave directory and run

$ pip install -r requirements.txt
$ pip install .

Installing using Docker

Another straightforward installation which does not depend on specific compilers relies on docker virtualisation system.

To install the docker image and test it is working:

$ docker pull pyreefmodel/radwave:latest
$ docker run --rm pyreefmodel/radwave:latest help

On Linux, to build the dockerfile locally, we provide a script. First ensure you have checked out the source code from github and then run the script in the Docker directory. If you modify the dockerfile and want to push the image to make it publicly available, it will need to be retagged to upload somewhere other than the pyReef-model repository.

$ git checkout https://github.com/pyReef-model/RADWave.git
$ cd RADWave
$ source Docker/build-dockerfile.sh

Note

For non-Linux platforms, the use of Docker Desktop for Mac or Docker Desktop for Windows is recommended. The docker container to look for is named pyreefmodel/radwave!