TRACK-pylib 0.4.5

pyTRACK

TRACK ^{1 2 3 4} is a powerful storm tracking software package that automatically identifies and tracks storm features in model and observational data. pyTRACK is intended to be an implementation of TRACK on Python that ports over most features of TRACK, while being easy to install and use.

If you're on a Linux-based system, pyTRACK can be installed by simply running (Best to use a conda environment with Python>=3.9)

pip install track-pylib

If that doesn't work, git clone the stable branch of this repository and pip install from the base directory.

git clone -b stable https://github.com/Ai33L/pyTRACK.git
cd pyTRACK
pip install -e .

The 'stable' branch contains the latest tested code, and the 'main' branch is used actively for development.

Then from a Python terminal anywhere, run

from pyTRACK import *
track()

This should start the TRACK namelist and should behave exactly like if you ran bin/track.linux from the compiled TRACK folder. The input and output files are assumed to be at the current working directory.

Running track() should work without any additional packages. However, some other pyTRACK functionalities depend on having cdo and nco installed on the system. You will be prompted to install these if you don't have them already. The easiest way to do this is work on a conda environment and run

conda install conda-forge::python-cdo
conda install conda-forge::pynco

pyTRACK also supports some pre-set workflows, and is under active development. To see a list of workflows currently available, and for a more extensive documentation, check out here.