This repository contains three notebooks that can be used to recreate the scientific figures of the PNAS paper describing mrCOSTS. The data used are included. These data were altered slightly from their original versions for ease of use (i.e., subsets, converting to xarray friendly formats).
- numpy
- scipy
- matplotlib
- seaborn
- xarray
- pydmd
- cartopy
- netcdf4
- h5netcdf
You might have to do some trouble shooting. This repository is provided as a demonstration, not a tutorial.
You will need to create a directory called "data" by expanding data.zip.
Next, you will need to actually fit mrCOSTS by setting the fit flag to True in the desired notebook. After fitting, the model results will be stored and can be analyzed without fitting again by setting fit back to False.