SVMmulticlass Multi-Class Support Vector Machine Author: Thorsten Joachims <[email protected]> Cornell University Department of Computer Science Version: 2.20 Date: 14.08.2008 Overview SVMmulticlass uses the multi-class formulation described in [1], but optimizes it with an algorithm that is very fast in the linear case. For a training set (x1,y1) ... (xn,yn) with labels yi in [1..k], it finds
{{#tags}}- {{label}}
{{/tags}}