2007-07-28 09:09:40 +02:00
|
|
|
LIBLINEAR is a linear classifier for data with millions of instances and
|
2020-06-14 16:22:35 +02:00
|
|
|
features. It supports:
|
|
|
|
|
- L2-regularized classifiers
|
|
|
|
|
- L2-loss linear SVM, L1-loss linear SVM, and logistic regression (LR)
|
|
|
|
|
- L1-regularized classifiers (after version 1.4)
|
|
|
|
|
- L2-loss linear SVM and logistic regression (LR)
|
|
|
|
|
- L2-regularized support vector regression (after version 1.9)
|
|
|
|
|
- L2-loss linear SVR and L1-loss linear SVR.
|
2007-05-27 16:05:54 +02:00
|
|
|
|
2020-06-14 16:22:35 +02:00
|
|
|
Main features of LIBLINEAR include:
|
|
|
|
|
- Same data format as LIBSVM, our general-purpose SVM solver, and also similar
|
|
|
|
|
usage
|
|
|
|
|
- Multi-class classification: 1) one-vs-the rest, 2) Crammer & Singer
|
|
|
|
|
- Cross validation for model evaulation
|
|
|
|
|
- Automatic parameter selection
|
2008-06-03 04:29:39 +02:00
|
|
|
- Probability estimates (logistic regression only)
|
|
|
|
|
- Weights for unbalanced data
|
2020-06-14 16:22:35 +02:00
|
|
|
- MATLAB/Octave, Java, Python, Ruby interfaces
|
2007-05-27 16:05:54 +02:00
|
|
|
|
2020-06-14 16:22:35 +02:00
|
|
|
WWW: https://www.csie.ntu.edu.tw/~cjlin/liblinear/
|
