Age | Commit message (Collapse) | Author | Files | Lines |
|
Version 2.41 released on July 29, 2020 (some bug fixes of version 2.40).
Version 2.40 released on July 22, 2020.
A new solver: dual coordinate descent method for linear one-class SVM; see the paper
The Newton solver is updated to have faster training speed; see the release note
A new option -R to allow users not to regularize bias (when -B 1 is used)
|
|
|
|
This notably fixes building with RELRO enabled.
Bump PKGREVISION, since this generates a different binary now that SSP and
FORTIFY are enabled.
|
|
We have improved the trust-region update rule in the primal-based Newton method. It's significantly faster (e.g., twice faster or more) on some problems (see the technical report).
We now support scipy objects in the Python interface
|
|
|
|
Unknown
|
|
Problems found locating distfiles:
Package dfftpack: missing distfile dfftpack-20001209.tar.gz
Package eispack: missing distfile eispack-20001130.tar.gz
Package fftpack: missing distfile fftpack-20001130.tar.gz
Package linpack: missing distfile linpack-20010510.tar.gz
Package minpack: missing distfile minpack-20001130.tar.gz
Package odepack: missing distfile odepack-20001130.tar.gz
Package py-networkx: missing distfile networkx-1.10.tar.gz
Package py-sympy: missing distfile sympy-0.7.6.1.tar.gz
Package quadpack: missing distfile quadpack-20001130.tar.gz
Otherwise, existing SHA1 digests verified and found to be the same on
the machine holding the existing distfiles (morden). All existing
SHA1 digests retained for now as an audit trail.
|
|
LIBLINEAR is a linear classifier for data with millions of instances
and 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.
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 selection
Probability estimates (logistic regression only)
Weights for unbalanced data
MATLAB/Octave, Java, Python, Ruby interfaces
|