https://github.com/caselles/entropic_affinities
A sparsification method for graph-based clustering and semi-supervised learning tasks, implemented on Python.
https://github.com/caselles/entropic_affinities
Last synced: 21 days ago
JSON representation
A sparsification method for graph-based clustering and semi-supervised learning tasks, implemented on Python.
- Host: GitHub
- URL: https://github.com/caselles/entropic_affinities
- Owner: Caselles
- Created: 2016-12-25T17:53:35.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2017-07-04T18:50:37.000Z (about 9 years ago)
- Last Synced: 2025-03-29T17:14:17.402Z (over 1 year ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 727 KB
- Stars: 1
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Entropic affinities
Author : Hugo Caselles-Dupré
Disclaimer : several syntax errors can exist in the code. Use it wisely.
---------------------------------------
A sparsification method for graph-based clustering and semi-supervised learning tasks.
---------------------------------------
In this directory you can find a Jupyter Notebook with the implementation of Entropic Affinities, following this article : http://jmlr.org/proceedings/papers/v28/vladymyrov13.pdf.
Entropic affinities is a method to automatically tune the bandwith of each data point when creating a sparse graph on a data set, with a kernel-based method. Here, the implementation uses the Gaussian kernel.
------------------------------------------
If you have any questions : casellesdupre.hugo@gmail.com.