The package contains HingeBoost for binary and multi-class classification, with unequal misclassification costs for binary case. Functional gradient descent algorithm to optimize the hinge loss. The algorithm can fit linear and nonlinear classifiers.
| Version: | 0.3-2 |
| Depends: | rpart, gbm |
| Published: | 2012-01-17 |
| Author: | Zhu Wang |
| Maintainer: | Zhu Wang <zwang at ccmckids.org> |
| License: | GPL-2 |
| In views: | MachineLearning |
| CRAN checks: | bst results |
| Package source: | bst_0.3-2.tar.gz |
| MacOS X binary: | bst_0.3-2.tgz |
| Windows binary: | bst_0.3-2.zip |
| Reference manual: | bst.pdf |
| News/ChangeLog: | NEWS |
| Old sources: | bst archive |
| Reverse suggests: | caret |