freebsd-ports/science/liblinear/pkg-descr

16 lines
562 B
Text

LIBLR is a linear classifier for data with millions of instances and
features. It implement a trust region Newton method in
C.-J. Lin, R. C. Weng, and S. S. Keerthi. Trust region Newton method
for large-scale regularized logistic regression. Technical report, 2007.
A short version appears in ICML 2007.
Main features of LIBLR include
Same data format as LIBSVM and similar usage
One-vs-the rest multi-class classification
Cross validation for model selection
Probability estimates
Weights for unbalanced data
WWW: http://www.csie.ntu.edu.tw/~cjlin/liblr/