LiquidSVM Versions Save

Support vector machines (SVMs) and related kernel-based learning algorithms are a well-known class of machine learning algorithms, for non-parametric classification and regression. liquidSVM is an implementation of SVMs whose key features are: fully integrated hyper-parameter selection, extreme speed on both small and large data sets, full flexibility for experts, and inclusion of a variety of different learning scenarios: multi-class classification, ROC, and Neyman-Pearson learning, and least-squares, quantile, and expectile regression.

CRAN-v1.2.0

6 years ago

liquidSVM v1.2.0 on CRAN (Release date: 2017-07-15)

  • version now up to date with core version
  • added predict.prob parameter to activate conditional probability estimation
  • added grouped cross validation
  • mlr support
  • added more explicit aliases for learning scenarios: svmRegression for lsSVM, svmMulticlass for mcSVM, svmQuantileRegression for qtSVM, and svmExpectileRegression for exSVM. The old ones are still valid and the main implementation.
  • added configuration defaults, e.g.: options(liquidSVM.default.display=1), options(liquidSVM.default.scale=TRUE), ...
  • predict now returns the correct number of columns for expectile, quantile, ...
  • fixed demo vignette run time issues
  • fixed CUDA-compilation issue
  • TARGET defaults to "default" on Sparc
  • fixed PROTECT-issues (thanks to kalibera/rchk!) and switched from CXX1X to CXX11
  • test-coverage over 90%