GraphDiffusion is a python package based on the DiffusionGeometry library in Matlab (https://services.math.duke.edu/~mauro/code.html#DiffusionGeom).

1. GraphDiffusion has been implemented in Python3 and can be installed using

    $> git clone git@github.com:pkathail/GraphDiffusion.git
    $> cd GraphDiffusion
    $> sudo pip3 install .
   

2. GraphDiffusion depends on a number of python3 packages available on pypi and these dependencies are listed in setup.py All the dependencies will be automatrically installed using the above commands

After installation, GraphDiffusion can be used with the following commands

	$> import GraphDiffusion
	$> res = GraphDiffusion.graph_diffusion.run_diffusion_map(data, knn=10, normalization='smarkov')

where data is a N x D matrix representing N points in R ^ D, knn is the number of nearest neighbors, normalization is the method for normalizing weights. Please refer to the docstring for more details.

res is dictionary with the following objects:

1. T: N x N sparse matrix giving the normalized diffusion operator
2. W: N x N sparse matrix of weights
3. EigenVectors: Eigen vectors of matrix T
4. EigenValues: Eigen values of matrix T