-------------------
Version: 4.44
Date: 2010-07-19

Added neuralnet, quantregForest and rda:::rda to train(). 
Since there is a naming conflict with mda:::rda, the 
rda:::rda model was given a method value of "scrda".

-------------------
Version: 4.43
Date: 2010-06-30

Bug fix release:

o the resampling estimate of the standard deviation given 
  by train() since v 4.39 was wrong

o a new field was added to varImp.mvr called "estimate".
  In cases where the mvr model had multiple estimates of
  performance (e.g. training set, CV, etc) the user can
  now select which estimate they want to be used in the
  importance calculation (thanks to Sophie Brand for
  finding this)

-------------------
Version: 4.42
Date: 2010-06-09

Added predict.sbf and modified the structure of the sbf 
helper functions. The "score" function only computes 
the metric used to filter and the filter function does
the actual filtering. This was changed so that FDR 
corrections or other operations that use all of the 
p-values can be computed. 

Also, the formatting of p-values in print.confusionMatrix  
was changed and an argument was added to maxDissim
so that the variable name is returned instead of the index.

Independent component analysis was added to the list of 
pre-processing operations and a new model ("icr") was
added to fit a pcr-like model with the ICA components.

-------------------
Version: 4.40
Date: 2010-05-19

Added hda and cleaned up the caret training vignette 

-------------------
Version: 4.39
Date: 2010-05-15

Added several classes for examining the resampling results. There
are methods for estimating pair-wise differences and lattice
functions for visualization. The training vignette has a new
section describing the new features.

-------------------
Version: 4.38
Date: 2010-05-12

Added partDSA and stepAIC for linear models and generalized
linear models

-------------------
Version: 4.37
Date: 2010-04-18

Fixed a new bug in how resampling results are exported

-------------------
Version: 4.36
Date: 2010-04-17

Added penalized linear models from the foba package

-------------------
Version: 4.35
Date: 2010-04-15

Added rocc classification and fixed a typo.

-------------------
Version: 4.34
Date: 2010-03-17

Added two new data sets: dhfr and cars

-------------------
Version: 4.33
Date: 2010-03-01

Added GAMens (ensembles using gams)

Fixed a bug in roc() that, for some data cases, would reverse the "positive"
class and report sensitivity as specificity and vice-versa.

-------------------
Version: 4.32
Date: 2009-12-24

Added a parallel random forest method in train() using the foreach package. 
Also added penalized logistic regression using the plr() function in the
stepPlr package.

-------------------
Version: 4.31
Date: 2009-12-07

Added a new feature selection function, sbf (for selection by filter).

Fixed bug in rfe that did not affect the results, but did produce
a warning.

A new model function, nullModel, was added. This model fits either the
mean only model for regression or the majority class model for classification.

Also, ldaFuncs had a bug fixed.

Minor changes to Rd files

-------------------
Version: 4.30
Date: 2009-11-09

For whatever reason, there is now a function in the spls package
by the name of splsda that does the same thing. A few functions
and a man page were changed to ensure backwards compatibility.

-------------------
Version: 4.29
Date: 2009-11-06

Added stepwise variable selection for lda and qda using the 
stepclass function in klaR

-------------------
Version: 4.28
Date: 2009-11-05

Added robust linear and quadratic discriminant analysis functions 
from rrcov.

Also added another column to the output of extractPRob and 
extractPrediction that saves the name of the model object so
that you can have multiple models of the same type and tell
which predictions came from which model.

Changes were made to plotClassProbs: new parameters were added
and densityplots can now be produced.

-------------------
Version: 4.27
Date: 2009-11-01

Added nodeHarvest


-------------------
Version: 4.26
Date: 2009-10-26

Fixed a bug in caretFunc() that led to NaN variable rankings, so that
the first k terms were always selected.


-------------------
Version: 4.25
Date: 2009-10-03

Added parallel processing functionality for rfe()


-------------------
Version: 4.24
Date: 2009-09-29

Added the ability to use custom metrics with rfe()


-------------------
Version: 4.22
Date: 2009-09-23

Many Rd changes to work with updated parser.


-------------------
Version: 4.21
Date: 2009-09-18

Re-saved data in more compressed format

-------------------
Version: 4.20
Date: 2009-07-19

Added pcr as a method


-------------------
Version: 4.19
Date: 2009-06-30

Weights argument was added to train for models that accept weights

Also, a bug was fixed for lasso regression (wrong lambda 
specification) and other for prediction in naive Bayes models
with a single predictor.


-------------------
Version: 4.18
Date: 2009-06-17

Fixed bug in new nearZeroVar and updated format.earth so that it
does not automatically print the formula


-------------------
Version: 4.17
Date: 2009-06-04

Added a new version of nearZeroVar from Allan Engelhardt that is
much faster


-------------------
Version: 4.16
Date: 2009-05-21

Fixed bugs in extractProb (for glmnet) and filterVarImp.

For glmnet, the user can now pass in their own value of family to 
train (otherwise train will set it depending on the mode of the 
outcome). However, glmnet doesn't have much support for families at
this time, so you can't change links or try other distributions. 

-------------------
Version: 4.15
Date: 2009-05-13

Fixed bug in createFolds when the smallest y value is more than 25%
of the data

-------------------
Version: 4.14
Date: 2009-05-13

Fixed bug in print.train

-------------------
Version: 4.13
Date: 2009-05-12

Added vbmp from vbmp package

-------------------
Version: 4.12
Date: 2009-05-07

Added additional error check to confusionMatrix and fixed an
absurd typo in print.confusionMatrix

-------------------
Version: 4.11
Date: 2009-04-25

Added 

  - linear kernels for svm, rvm and Gaussian processes
  - rlm from MASS
  - a knn regression model, knnreg
  - a set of functions (class "classDist") to computes the class 
    centroids and covariance matrix for a training set for 
    determining Mahalanobis distances of samples to each class 
    centroid 
  - a set of functions (rfe) for doing recursive feature selection 
    (aka backwards selection). A new vignette was added for more 
    details

-------------------
Version: 4.10
Date: 2009-03-20

Added OneR and PART from RWeka

-------------------
Version: 4.09
Date: 2009-03-20

- Fixed error in documentation for confusionMatrix:

  old:     Detection Prevalence = \frac{A}{A+B}
  correct: Detection Prevalence = \frac{A+B}{A+B+C+D}

  The underlying code was correct.

- Added lars (fraction and step as parameters)

-------------------
Version: 4.08
Date: 2009-02-18

Updated train and bagEarth to allow earth for classification models

-------------------
Version: 4.07
Date: 2009-01-25

Added glmnet models

-------------------
Version: 4.06
Date: 2009-01-24

Added code for sparse PLS classification.

Fix a bug in prediction for caTools::LogitBoost

-------------------
Version: 4.05
Date: 2009-01-23

Updated again for more stringent R CMD check tests in R-devel 2.9

-------------------
Version: 4.04
Date: 2009-01-22

Updated for more stringent R CMD check tests in R-devel 2.9

-------------------
Version: 4.03
Date: 2009-01-20

Significant internal changes were made to how the models are fit. Now, the 
function used to compute the models is passed in as a parameter (defaulting to 
lapply). In this way, users can use their own parallel processing software 
without new versions of caret. Examples are given in ?train.

Also, fixed a bug where the MSE (instead of RMSE) was reported for 
random forest OOB resampling

There are more examples in ?train.

Changes to confusionMatrix, sensitivity, specificity and the 
predictive value functions:

  - each was made more generic with default and table methods
  - confusionMatrix "extractor" functions for matrices and tables 
    were added
  - the pos/neg predicted value computations were changed to incorporate 
    prevalence
  - prevalence was added as an option to several functions 
  - detection rate and prevalence statistics were added to confusionMatrix
  - the examples were expanded in the help files

This version of caret will break compatibility with caretLSF and 
caretNWS. However, these packages will not be needed now and will be 
deprecated.

-------------------
Version: 3.51
Date: 2008-12-03

Updated the man files and manuals.

-------------------
Version: 3.50
Date: 2008-12-02

Added qda, mda and pda.

-------------------
Version: 3.49
Date: 2008-11-30

Fixed bug in resampleHist. Also added a check in the train functions
that error trapped with glm models and > 2 classes

-------------------
Version: 3.48
Date: 2008-11-30

Added glms. Also, added varImp.bagEarth to the namespace.

-------------------
Version: 3.47
Date: 2008-11-24

Added sda from the sda package. There was a naming conflict between 
sda::sda and sparseLDA:::sda. The method value for sparseLDA was
changed from "sda" to "sparseLDA".

-------------------
Version: 3.46
Date: 2008-11-11

Added spls from the spls package

-------------------
Version: 3.45
Date: 2008-10-17

Added caching of RWeka objects to that they can be saved to the file system and
used in other sessions. (changes per Kurt Hornik on 2008-10-05)

-------------------
Version: 3.44
Date: 2008-10-15

Added sda from the sparseLDA package (not on CRAN). 

Also, a bug was fixed where the ellipses were not passed into a few of the newer 
models (such as penalized and ppr)

-------------------
Version: 3.43
Date: 2008-10-03

Added the penalized model from the penalized package. In caret, it is regression only
although the package allows for classification via glm models. However, it does not
allow the user to pass the classes in (just an indicator matrix). Because of this, it
doesn't really work with the rest of the classification tools in the package.

-------------------
Version: 3.42
Date: 2008-09-26

Added a little more formatting to print.train

-------------------
Version: 3.41
Date: 2008-09-19

For gbm, let the user over-ride the default value of the distribution argument (brought
us by Peter Tait via RHelp).

-------------------
Version: 3.40
Date: 2008-09-18

Changed predict.preProcess so that it doesn't crash if newdata does not have all of the
variables used to originally pre-process *unless* PCA processing was requested.

-------------------
Version: 3.39
Date: 2008-09-18

Fixed bug in varImp.rpart when the model had only primary splits.

Minor changes to the Affy normalization code

Changed typo in predictors man page

-------------------
Version: 3.38
Date: 2008-09-09

Added a new class called predictors that returns the names of the predictors that were
used in the final model.

Also added ppr from the stats package.

Minor update to the project web page to deal with IE issues

-------------------
Version: 3.37
Date: 2008-09-04

Added the ability of train to use custom made performance functions so that the tuning
parameters can be chosen on the basis of things other than RMSE/R-squared and 
Accuracy/Kappa. Specific changes:

  - a new argument was added to trainControl called "summaryFunction" that is used
    to specify the function used to compute performance metrics. The default function
    preserves the functionality prior to this new version

   - a new argument to train is "maximize" which is a logical for whether the 
    performance measure specified in the "metric" argument to train should be 
    maximized or minimized.

   - the selection function specified in trainControl carries the maximize argument with
     it so that customized performance metrics can be used.

Other changes:

  - a bug was fixed in confusionMatrix (thanks to Gabor Grothendieck)

  - another bug was fixed related to predictions from least square SVMs 

-------------------
Version: 3.36
Date: 2008-08-29

Added superpc from the superpc package. One note: the data argument that is passed
to superpc is saved in the object that results from superpc.train. This is used later in
the prediction function.

-------------------
Version: 3.35
Date: 2008-08-27

Added slda from ipred

-------------------
Version: 3.34
Date: 2008-08-25

Fixed a few bugs related to the lattice plots from version 3.33.

Also added the ripper (aka JRip) and logistic model trees from RWeka

-------------------
Version: 3.33
Date: 2008-08-22

Added xyplot.train, densityplot.train, histogram.train and stripplot.train. These are 
all functions to plot the resampling points. There is some overlap between these
functions, plot.train and resampleHist. plot.train gives the average metrics only while 
these plot all of the resampled performance metrics. resampleHist could plot all of the 
points, but only for the final optimal set of predictors.

To use these functions, there is a new argument in trainControl called returnResamp 
which should have values "none", "final" and "all". The default is "final" to be consistent 
with previous versions, but "all" should be specified to use these new functions to their 
fullest.

-------------------
Version: 3.32
Date: 2008-07-28

The functions "predict.train" and "predict.list" were added to use as alternatives to the
extractPrediction and extractProbs functions. 

Added C4.5 (aka J48) and rules-based models (M5 prime) from RWeka.

Also added logitBoost from the caTools package. This package doesn't have a 
namespace and RWeka has a function with the same name. It was suggested to use the
"::" prefix to differentiate them (but we'll see how this works).



