LIBLINEAR is a linear classifier for data with millions of instances and features. It supports L2-regularized logistic regression (LR), L2-loss linear SVM, L1-loss linear SVM, and multi-class SVM from Crammer and Singer. The approach for L1-SVM, L2-SVM, and multi-class SVM from Crammer and Singer is a dual coordinate descent method. We implement a trust region Newton method for LR and L2-SVM.
- Changes to previous version:
Initial Announcement on mloss.org.
- BibTeX Entry: Download
- Corresponding Paper BibTeX Entry: Download
- URL: Project Homepage
- JMLR MLOSS PaperURL: JMLR-MLOSS Paper Homepage
- Supported Operating Systems: Cygwin, Linux, Windows, Macos
- Data Formats: None
- Tags: Svm, Classification, Support Vector Machines, Linear Svm, Large Scale Learning
- Archive: download here
Leave a comment
You must be logged in to post comments.