About: This package is an implementation of a linear svm solver with a wide class of regularizations on the svm weight vector (l1, l2, mixed norm l1-lq, adaptive lasso). We provide solvers for the classical single task svm problem and for multi-task with joint feature selection or similarity promoting term.

Initial Announcement on mloss.org.

About: A stochastic variant of the mirror descent algorithm employing Langford and Zhang's truncated gradient idea to minimize L1 regularized loss minimization problems for classification and regression.

Fixed major bug in implementation. The components of the iterate where the current example vector is zero were not being updated correctly. Thanks to Jonathan Chang for pointing out the error to us.

About: The Universal Java Matrix Package (UJMP) is a data processing tool for Java. Unlike JAMA and Colt, it supports multi-threading and is therefore much faster on current hardware. It does not only support matrices with double values, but instead handles every type of data as a matrix through a common interface, e.g. CSV files, Excel files, images, WAVE audio files, tables in SQL data bases, and much more.

Initial Announcement on mloss.org.

About: A (randomized) coordinate descent procedure to minimize L1 regularized loss for classification and regression purposes.

Fixed some I/O bugs. Lines that ended with whitespace were not read correctly in the previous version.