trainNWS {caretNWS}R Documentation

Fit Predictive Models over Different Tuning Parameters

Description

This function sets up a grid of tuning parameters for a number of classification and regression routines, fits each model and calculates a resampling based performance measure.

Usage

trainNWS(x, y, method = "rf",  ...,
   metric = ifelse(is.factor(y), "Accuracy", "RMSE"),
      maximize = ifelse(metric == "RMSE", FALSE, TRUE),
   sleighObj = NULL,
   trControl = trainNWSControl(), 
   tuneGrid = NULL,
   tuneLength = 3)

Arguments

x a data frame containing training data where samples are in rows and features are in columns.
y a numeric or factor vector containing the outcome for each sample.
method a string specifying which classification or regression model to use. Possible values are: lm, rda, lda, gbm, rf, nnet, multinom, gpls, lvq, rpart, knn, pls, pam, nb, earth, treebag, svmpoly, svmradial, fda, bagEarth, bagFDA, glmboost, gamboost, blackboost, ada, ctree, cforest, sddaLDA, sddaQDA, LogitBoost, J48, M5Rules, slda, superpc, enet and lasso. See the Details section below.
... arguments passed to the classification or regression routine (such as randomForest). Errors will occur if values for tuning parameters are passed here.
metric a string that specifies what summary metric will be used to select the optimal model. Possible values are "RMSE" and "Rsquared" for regression and "Accuracy" and "Kappa" for classification.(NOTE: If given, this argument must be named.)
maximize a logical: should the metric be maximized or minimized?
sleighObj a sleigh object for an existing nws session. This object is not destroyed after trainNWS closes.
trControl a list of values that define how this function acts. See trainNWSControl. (NOTE: If given, this argument must be named.)
tuneGrid a data frame with possible tuning values. The columns are named the same as the tuning parameters in each method preceded by a period (e.g. .decay, .lambda). See the function createGrid in this package for more details. (NOTE: If given, this argument must be named.)
tuneLength an integer denoting the number of levels for each tuning parameters that should be generated by createGrid. (NOTE: If given, this argument must be named.)

Details

trainNWS can be used to tune models by picking the complexity parameters that are associated with the optimal resampling statistics. For particular model, a grid of parameters (if any) is created and the model is trained on slightly different data for each candidate combination of tuning parameters. Across each data set, the performance of held-out samples is calculated and the mean and standard deviation is summarized for each combination. The combination with the optimal resampling statistic is chosen as the final model and the entire training set is used to fit a final model.

Currently, the trainNWS function does not support model specification via a formula. It assumes that all of the predictors are numeric (perhaps generated by model.matrix).

A variety of models are currently available. The table below enumerates the models and the values of the method argument, as well as the complexity parameters used by trainNWS.

Model method Value Package Tuning Parameter(s)
Recursive partitioning rpart rpart maxdepth
ctree party mincriterion
Boosted Trees gbm gbm interaction depth,
n.trees, shrinkage
blackboost mboost maxdepth, mstop
ada ada maxdepth, iter, nu
Boosted regression models glmboost mboost mstop
gamboost mboost mstop
logitboost caTools nIter
Random forests rf randomForest mtry
cforest party mtry
Bagged Trees treebag ipred None
Neural networks nnet nnet decay, size
Partial least squares pls pls, caret ncomp
Support Vector Machines (RBF) svmradial kernlab sigma, C
Support Vector Machines (polynomial) svmpoly kernlab scale, degree, C
Relevance Vector Machines (RBF) rvmradial kernlab sigma
Relevance Vector Machines (polynomial) rvmpoly kernlab scale, degree
Least Squares Support Vector Machines (RBF) lssvmradial kernlab sigma
Gaussian Processes (RBF) guassprRadial kernlab sigma
Gaussian Processes (polynomial) guassprPoly kernlab scale, degree
Linear least squares lm stats None
Multivariate adaptive regression splines earth earth degree, nprune
Bagged MARS bagEarth caret, earth degree, nprune
M5 Rules M5Rules RWeka pruned
Elastic Net enet elasticnet lambda, fraction
The Lasso enet elasticnet fraction
Supervised Principal Components superpc superpc n.components, threshold
Linear discriminant analysis lda MASS None
Stabilised Linear discriminant analysis slda ipred None
Stepwise Diagonal Discriminant Analysis sddaLDA, sddaQDA SDDA None
Logistic/multinomial regression multinom nnet decay
Regularized discriminant analysis rda klaR lambda, gamma
Stabilised Linear Discriminant Analysis slda ipred None
Flexible discriminant analysis (MARS) fda mda, earth degree, nprune
Bagged FDA bagFDA caret, earth degree, nprune
C4.5 decision trees J48 RWeka C
k nearest neighbors knn3 caret k
Nearest shrunken centroids pam pamr threshold
Naive Bayes nb klaR usekernel
Generalized partial least squares gpls gpls K.prov
Learned vector quantization lvq class k

By default, the function createGrid is used to define the candidate values of the tuning parameters. The user can also specify their own. To do this, a data fame is created with columns for each tuning parameter in the model. The column names must be the same as those listed in the table above with a leading dot. For example, ncomp would have the column heading .ncomp. This data frame can then be passed to createGrid.

In some cases, models may require control arguments. These can be passed via the three dots argument. Note that some models can specify tuning parameters in the control objects. If specified, these values will be superseded by those given in the createGrid argument.

The vignette entitled "caret Manual – Model Building" in the caret package has more details and examples related to this function.

caretNWS is a parallel version of the train function in the caret package. This function implements a combination of sequential and parallel jobs. For example, if a PLS model with 10 candidate values for the number of retained components is requested along with 10-fold cross–validation, the 10 jobs for each caudate tuning parameter are submitted in parallel. Once finished, the process is started again with the second candidate tuning parameter etc.

The function also has a fault-tolerance function where any jobs that are abnormally long (say 20x the longest job to date) can be killed automatically. In this way, a problem with a worker node may not destroy the results to date. For example, if the models are tuned using 25 bootstrap results, the failure of one node will cause the function to tune the model with the remainder and keep going.

The nwsStart can be dependent on the configuration of the user's system. The example function given in the package may not work out of the box.

Value

A list is returned of class train containing:

modelType an identifier of the model type.
results a data frame the training error rate and values of the tuning parameters.
call the (matched) function call with dots expanded
dots a list containing any ... values passed to the original call
metric a string that specifies what summary metric will be used to select the optimal model.
trControl the list of control parameters.
finalModel an fit object using the best parameters
trainingData a data frame
resample A data frame with columns for each performance metric. Each row corresponds to each resample. If leave-one-out cross-validation or out-of-bag estimation methods are requested, this will be NULL. The returnResamp argument of trainNWSControl
perfNames a character vector of performance metrics that are produced by the summary function
maximize a logical recycled from the function arguments.

Author(s)

Max Kuhn

See Also

train, createGrid, createFolds

Examples

## Not run: 
data(iris)
TrainData <- iris[,1:4]
TrainClasses <- iris[,5]

knnFit <- trainNWS(TrainData, TrainClasses, "knn", tuneLength = 10,
   trControl = trainNWSControl(method = "boot"))
## End(Not run)

[Package caretNWS version 0.23 Index]