Skip to content

PersiGraph is a package specialized in clustering ensemble of time-series, notably when the ensemble size is small such as in ensemble weather prediction

License

Notifications You must be signed in to change notification settings

nglm/persigraph

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PersiGraph

PersiGraph is python a package specialized in clustering ensemble of time-series, notably when the ensemble size is small such as in ensemble weather prediction.

PersiGraph generates a graph where the vertices represent the different clusters and the edges the evolution of the clusters with time. It allows for a dynamic number of cluster, meaning that clusters can merge and split with time.

In addition to a fully automated generation of the most relevant graph of clusters, PersiGraph aims at taking into account the potential uncertainty around the number of clusters when building the graph, so that the user can get a more faithful representation of the data and can make an informed decision on the final number of clusters at each time step. To assess the quality of the clusters, PersiGraph relies on the package PyCVI.

Install

In a poetry project, run:

poetry add git+https://github.com/nglm/persigraph.git

About

PersiGraph is a package specialized in clustering ensemble of time-series, notably when the ensemble size is small such as in ensemble weather prediction

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages