[R-pkgs] New caret packages

[Max]

New major versions of the caret packages (caret 3.37, caretLSF 1.23 and
caretNWS 0.23) have been uploaded to CRAN.

caret is a package for building and evaluating a wide variety of predictive
models. There are functions for pre-processing, tuning models using
resampling, visualizing the results, calculating performance and estimating
variable importance.  caretNWS and caretLSF are two parallel processing
versions that can reduce the training time when multiple compute nodes are
available.

The project is now hosted on R-Forge. The homepage is

   http://caret.r-forge.r-project.org/

The package currently includes model tuning/resampling for the following
models: lm, single trees (C4.5, rpart, ctree, logistic model trees), mars
(via earth), boosted models (ada, gbm, blackboost, glmboost, gamboost,
logitboost), bagged models (trees, earth, fda), randomforests (randomforest
and cforest), rule-based models (Ripper and M5 prime), discriminant models
(lda, fda, rda, ssda, slda), kernel methods (lssvm, ksvm, rvm, gausspr),
nnet, nnet with initial pca step, multinom, pls, plsda, gpls, nearest
shrunken centroids, the lasso, the elastic net, supervised pca, knn, lvq and
NaiveBayes.

Recent changes include:
 - Estimation of class probabilities from PLS discriminant analysis using
Bayes rule (in addition to softmax)
- Added predict.train and predit.list
- More lattice plots to visualize resampling results (xyplot, stripplot,
densitplot, histogram)
- User-specified performance metrics for resampling
- User-specified algorithms for determining the optimal tuning parameters
(instead of highest/lowest)
- A CHANGES files now exists to track the specifics of the version changes
 
Max