A python tool for implementing the Autonomous Gaussian Decomposition algorithm.

Robert R. Lindner, Carlos Vera-Ciro, Claire E. Murray, Elijah Bernstein-Cooper

Lindner et al. 2015

You will need the following packages to run GaussPy. We list the version of each package which we know to be compatible with GaussPy.

• python 2.7

• numpy (v1.12.1)

• scipy (v0.17.0)

• lmfit (v0.9.3)

• h5py (v2.0.1)

If you do not already have Python 2.7, you can install the Anaconda Scientific Python distribution, which comes pre-loaded with numpy, scipy, and h5py.

If you wish to use GaussPy's plotting capabilities you will need to install matplotlib:

• matplotlib (v1.1.1)

If you wish to use optimization with Fortran code you will need

• GNU Scientific Library (GSL)

Download GaussPy using git $ git clone git:https://github.com/gausspy/gausspy.git

You will need several libraries which the GSL, h5py, and scipy libraries depend on. Install these required packages with:

sudo apt-get install libblas-dev liblapack-dev gfortran libgsl0-dev libhdf5-serial-dev 
sudo apt-get install hdf5-tools

Install pip for easy installation of python packages:

sudo apt-get install python-pip

Then install the required python packages:

sudo pip install scipy numpy h5py lmfit

Install the optional dependencies for plotting and optimization:

sudo pip install matplotlib
sudo apt-get install libgsl0-dev

Installation on OSX can be done easily with homebrew. Install pip for easy installation of python packages:

sudo easy_install pip

Then install the required python packages:

sudo pip install numpy scipy h5py lmfit

Install the optional dependencies for plotting and optimization:

sudo pip install matplotlib
sudo brew install gsl

To install make sure that all dependences are already installed and properly linked to python --python has to be able to load them--. Then cd to the local directory containing GaussPy and install via

python setup.py install

If you don't have root access and/or wish a local installation of GaussPy then use

python setup.py install --user

To contribute to GaussPy, see Contributing to GaussPy