epcreg {EnsemblePCReg}

Principal-Components-Regression-Based (PCR) Integration of Base Learners for Ensemble Learning

Description

This function applies PCR to predictions from regression base learners to produce an ensemble prediction. Number of PC's used in the PCR algorithm is determined by minimizing the cross-validation error. The data partition for the integration phase does not have to be the same as the partition(s) used to generate the base learners. Functions from EnsembleBase are used for training and prediction of base learners. Also, base classes and generic methods of the same package are extended to support PCR integration.

Usage

epcreg(formula, data
  , baselearner.control=epcreg.baselearner.control()
  , integrator.control=epcreg.integrator.control()
  , ncores=1, filemethod=FALSE, print.level=1
  , preschedule = TRUE
  , schedule.method = c("random", "as.is", "task.length")
  , task.length
)

Arguments

formula

Formula expressing response variable and covariates.

data

Data frame containing the response variable and covariates.

baselearner.control

Control structure determining the base learners, their configurations, and data partitioning details. See epcreg.baselearner.control.

integrator.control

Control structure governing integrator behavior. See epcreg.integrator.control.

ncores

Number of cores used for parallel training of base learners.

filemethod

Boolean flag indicating whether or not to save estimation objects to disk or not. Using filemethod=T reduces RAM pressure.

print.level

Controlling verbosity level.

preschedule

Boolean flag, indicating whether base learner training jobs must be scheduled statically (TRUE) or dynamically (FALSE).

schedule.method

Method used for scheduling tasks on threads. In "as.is" tasks are assigned to threads in a round-robin fashion for static scheduling. In dynamic scheduling, tasks form a queue without any re-ordering. In "random", tasks are first randomly shuffled, and the rest is similar to "as.is". In "task.length", a heuristic algorithm is used in static scheduling for assigning tasks to threads to minimize load imbalance, i.e. make total task lengths in threads roughly equal. In dynamic scheduling, tasks are sorted in descending order of expected length to form the task queue.

task.length

Vector of estimated task lengths, to be used in the "task.length" method of scheduling.

Value

An object of classes epcreg (if filemethod==TRUE, also has class of epcreg.file), a list with the following elements:

call

Copy of function call.

formula

Copy of formula argument in function call.

instance.list

An object of class Instance.List, containing all permutations of base learner configurations and random data partitions generated in the body of the function.

integrator.config

Copy of configuration object passed to the integrator. Object of class Regression.Integrator.PCR.SelMin.Config.

method

Integration method. Currently, only "default" is supported.

est

A list with these elements: 1) baselearner.cv.batch, an object of class Regression.CV.Batch.FitObj containing the fit object from CV batch training of base learners; 2) integrator, an object of class Regression.Integrator.PCR.SelMin.FitObj containing the fit object returned by the integrator.

y

Copy of response variable vector.

pred

Within-sample prediction of the ensemble model.

filemethod

Copy of passed-in filemethod argument.

Author(s)

Mansour T.A. Sharabiani, Alireza S. Mahani

See Also

epcreg.baselearner.control, epcreg.integrator.control, Instance.List, Regression.Integrator.PCR.SelMin.Config, Regression.CV.Batch.FitObj, Regression.Batch.FitObj, Regression.Integrator.PCR.SelMin.FitObj

Examples

data(servo)
myformula <- class~motor+screw+pgain+vgain
perc.train <- 0.7
index.train <- sample(1:nrow(servo), size = round(perc.train*nrow(servo)))
data.train <- servo[index.train,]
data.predict <- servo[-index.train,]
## to run longer test using all 5 default regression base learners
## try: est <- epcreg(myformula, data.train, ncores=2)
est <- epcreg(myformula, data.train, ncores=2
  , baselearner.control=epcreg.baselearner.control(
    baselearners="knn"
    , baselearner.configs = make.configs("knn"
      , config.df = expand.grid(kernel = "rectangular"
        , k = c(5, 10)))))
newpred <- predict(est, data.predict)