Skip to content

A way to plot high-dimensional decision boundaries via Voronoi tesselation onto 2D

License

Notifications You must be signed in to change notification settings

dainis-boumber/nd_boundary_plot

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

nd-boundary-plot

A method to draw decision boundaries.

Modes of operation:

  1. For 2D feature space and binary probabilistic classifier, probability surface is drawn

TODO: probability output for n_classes > 2

  1. For 2D feature space and a classifier that does not predict posteriors, hard boundary is shown.

TODO: when possible, use decision_function() method and project that onto surface.

  1. For multi-dimensional, a way to plot high-dimensional decision boundaries via Voronoi tesselation onto 2D. Based on work by Migut, G. and Worring, M. and Veenman, C. J.

TODO: implement alternative

Author: Dainis Boumber [email protected]

base code: https://stackoverflow.com/questions/37718347/plotting-decision-boundary-for-high-dimension-data

@Article{MigutDMKD2015, author = "Migut, G. and Worring, M. and Veenman, C. J.", title = "Visualizing Multi-Dimensional Decision Boundaries in 2D", journal = "Data Mining and Knowledge Discovery", year = "2015", url = "https://ivi.fnwi.uva.nl/isis/publications/2015/MigutDMKD2015", has_image = 1 }

About

A way to plot high-dimensional decision boundaries via Voronoi tesselation onto 2D

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages