Duplexity is a Python package that calculates validation metrics on meteorological data, including dynamical weather forecasts, AI model outputs and atmospheric reanalyses.
We have not yet made Duplexity pip-installable, as the project is still in the early development stages. This means the code is constantly changing, and we currently don't guarantee backwards compatibility when changes are made. Once a stable release of Duplexity is available, we will upload Duplexity to PyPI to make it pip-installable.
Ensure you have the following prerequisites installed on your system:
- Python (version 3.7 or later)
- Git
Navigate to the directory you would like to clone the Duplexity repository into, and clone from GitHub:
git clone https://github.com/lexixu19/duplexity.git
Create and/or activate the conda or pip environment you would like to use Duplexity within. For example:
conda create --name duplexity
conda activate duplexity
Ensure you are in the highest level of the Duplexity directory on your local system:
cd duplexity
You should be able to see pyproject.toml in this directory. Run the following command to install Duplexity in your environment:
pip install .
Note: if you are a contributor or editor of the Duplexity environment, you should use pip install -e . to allow you to make edits which are immediately reflected when you import locally.
If there are metrics you would like to see added to Duplexity, please get in touch with the development team: Lexi Xu or Emily O'Riordan.