Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Already on GitHub? Sign in to your account

Add support for SVDD (one-class SVM) #2807 #3013

Open
wants to merge 3 commits into
from

Conversation

Projects
None yet
3 participants

raynald commented Mar 28, 2014

Support Vector Data Description (SVDD) could be a nice enhancement to OneClassSVM implementation.

A technical implementation is described in this paper:
http://www.csie.ntu.edu.tw/~cjlin/papers/svdd.pdf

Source code compatible with libsvm is available here:
http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/#libsvm_for_svdd_and_finding_the_smallest_sphere_containing_all_data

I have modified the sklearn/svm/src/libsvm/svm.h and svm.cpp files to add additional functionality of SVM.

-s 5 SVDD
-s 6 R^2 L1SVM
-s 7 R^2 L2SVM

Owner

GaelVaroquaux commented Mar 28, 2014

I don't understand how this can be used from scikit-learn? It seems to me like this feature isn't finished.

Can you do an example that compares SVDD to standard one class SVM, please.

Coverage Status

Coverage remained the same when pulling 8a1c8db on raynald:GSoc14 into a3f81d9 on scikit-learn:master.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment