[R-pkgs] version 4.39 of the caret package
Max.Kuhn at pfizer.com
Mon May 17 22:17:41 CEST 2010
Version 4.39 of the caret package was sent to CRAN.
caret can be used to tune the parameters of predictive models using
resampling, estimate variable importance and visualize the results.
There are also various modeling and "helper" functions that can be
useful for training models. caret has wrappers to over 75 different
models for classification and regression. See the package vignettes or
the paper at
for more details. I'll also be giving a talk at this year's useR!
Since the last posting to this list:
- 23 additional models were added to train()
- weights can be passed in through train()
- feature selection methods have been added: recursive feature
elimination (rfe()) and selection by univariate filters (sbf()).
Both functions can be run in parallel.
- a set of functions (class "classDist") to computes the class
centroids and covariance matrix for a training set for
determining Mahalanobis distances of new samples to each class
- a faster version of nearZeroVar() due to Allan Engelhardt
- two new data sets were added
- several classes for examining the resampling results were added for
estimating pair-wise differences in models and lattice visualizations
The NEWS file has the blow-by-blow list of changes.
The package homepage is
Send questions, collaborations, comments etc to max.kuhn at pfizer.com.
More information about the R-packages