Home

The Grab 'em - Tag 'em - Graph 'em (GTG) algorithm

Grab 'em, Tag 'em, Graph 'em (GTG) is a feature detection, evolution and feature characterization algorithm. It was built for the weather/ climate application of identifying mesoscale convective complexes in highly resolved temporal and geospatial remote-sensed datasets and characterizing the features using various data sources. However, the application of it extends beyond meteorological realms as the GTG focus is on feature detection and evolution that leverages graph theory.

##What is needed? The GTG codebase is entirely Python. Quite a few scientific Python projects are used. The following python modules are needed. It is suggested to use Continuum Analytics Anaconda Python package to install all the Python modules painlessly.

• Python 2.7.x (We used Anaconda 2.3.0 64 bit libraries, which installs the following dependencies.)
• sciPy
• NumPy
• Networkx
• matplotlib

Other module dependences:

• netCDF4

• GrADS (We used OpenGrADS grads2 Version 2.0.1.oga.1).This is only needed if you need to convert a raw binary file to netCDF. - LATS4D A user of the code noted that if you are GrADS v.2.1.a2 your image output format will have to be changed from .gif to .png if you receive the following error "GXPRINT error: unsupported output format"

  Mac OS X 10.10 Installation Steps

  • Install Anaconda from Continuum Analytics
  • Install Homebrew
  • Install HDF5 using homebrew:

  brew tap homebrew/science
  brew install hdf5

  • Use pip to install netCDF4

  pip install netCDF4