About: MultiBoost is a multi-purpose boosting package implemented in C++. It is based on the multi-class/multi-task AdaBoost.MH algorithm [Schapire-Singer, 1999]. Basic base learners (stumps, trees, products, Haar filters for image processing) can be easily complemented by new data representations and the corresponding base learners, without interfering with the main boosting engine.

Major changes :

• The “early stopping” feature can now based on any metric output with the --outputinfo command line argument.

• Early stopping now works with --slowresume command line argument.

Minor fixes:

• More informative output when testing.

• Various compilation glitch with recent clang (OsX/Linux).

About: The open source Error-Correcting Output Codes (ECOC) library contains both state-of-the-art coding and decoding designs, as well as the option to include your own coding, decoding, and base classifier.

Initial Announcement on mloss.org.