gafsControl  R Documentation 
Control the computational nuances of the gafs
and
safs
functions
Many of these options are the same as those described for
trainControl
. More extensive documentation and examples
can be found on the caret website at
http://topepo.github.io/caret/featureselectionusinggeneticalgorithms.html#syntax and
http://topepo.github.io/caret/featureselectionusingsimulatedannealing.html#syntax.
The functions
component contains the information about how the model
should be fit and summarized. It also contains the elements needed for the
GA and SA modules (e.g. crossover, etc).
The elements of functions
that are the same for GAs and SAs are:
fit
, with arguments x
, y
, lev
,
last
, and ...
, is used to fit the classification or regression
model
pred
, with arguments object
and x
, predicts
new samples
fitness_intern
, with arguments object
,
x
, y
, maximize
, and p
, summarizes performance
for the internal estimates of fitness
fitness_extern
, with
arguments data
, lev
, and model
, summarizes performance
using the externally heldout samples
selectIter
, with
arguments x
, metric
, and maximize
, determines the best
search iteration for feature selection.
The elements of functions
specific to genetic algorithms are:
initial
, with arguments vars
, popSize
and ...
, creates an initial population.
selection
, with
arguments population
, fitness
, r
, q
, and
...
, conducts selection of individuals.
crossover
, with
arguments population
, fitness
, parents
and ...
,
control genetic reproduction.
mutation
, with arguments
population
, parent
and ...
, adds mutations.
The elements of functions
specific to simulated annealing are:
initial
, with arguments vars
, prob
, and
...
, creates the initial subset.
perturb
, with
arguments x
, vars
, and number
, makes incremental
changes to the subsets.
prob
, with arguments old
,
new
, and iteration
, computes the acceptance probabilities
The pages http://topepo.github.io/caret/featureselectionusinggeneticalgorithms.html and http://topepo.github.io/caret/featureselectionusingsimulatedannealing.html have more details about each of these functions.
holdout
can be used to hold out samples for computing the internal
fitness value. Note that this is independent of the external resampling
step. Suppose 10fold CV is being used. Within a resampling iteration,
holdout
can be used to sample an additional proportion of the 90%
resampled data to use for estimating fitness. This may not be a good idea
unless you have a very large training set and want to avoid an internal
resampling procedure to estimate fitness.
The search algorithms can be parallelized in several places:
each externally resampled GA or SA can be run independently
(controlled by the allowParallel
options)
within a GA, the
fitness calculations at a particular generation can be run in parallel over
the current set of individuals (see the genParallel
)
if inner resampling is used, these can be run in parallel (controls depend on the
function used. See, for example, trainControl
)
any parallelization of the individual model fits. This is also specific to the modeling function.
It is probably best to pick one of these areas for parallelization and the first is likely to produces the largest decrease in runtime since it is the least likely to incur multiple restarting of the worker processes. Keep in mind that if multiple levels of parallelization occur, this can effect the number of workers and the amount of memory required exponentially.
gafsControl( functions = NULL, method = "repeatedcv", metric = NULL, maximize = NULL, number = ifelse(grepl("cv", method), 10, 25), repeats = ifelse(grepl("cv", method), 1, 5), verbose = FALSE, returnResamp = "final", p = 0.75, index = NULL, indexOut = NULL, seeds = NULL, holdout = 0, genParallel = FALSE, allowParallel = TRUE ) safsControl( functions = NULL, method = "repeatedcv", metric = NULL, maximize = NULL, number = ifelse(grepl("cv", method), 10, 25), repeats = ifelse(grepl("cv", method), 1, 5), verbose = FALSE, returnResamp = "final", p = 0.75, index = NULL, indexOut = NULL, seeds = NULL, holdout = 0, improve = Inf, allowParallel = TRUE )
functions 
a list of functions for model fitting, prediction etc (see Details below) 
method 
The resampling method: 
metric 
a twoelement string that specifies what summary metric will
be used to select the optimal number of iterations from the external fitness
value and which metric should guide subset selection. If specified, this
vector should have names 
maximize 
a twoelement logical: should the metrics be maximized or
minimized? Like the 
number 
Either the number of folds or number of resampling iterations 
repeats 
For repeated kfold crossvalidation only: the number of complete sets of folds to compute 
verbose 
a logical for printing results 
returnResamp 
A character string indicating how much of the resampled summary metrics should be saved. Values can be “all” or “none” 
p 
For leavegroup out crossvalidation: the training percentage 
index 
a list with elements for each resampling iteration. Each list element is the sample rows used for training at that iteration. 
indexOut 
a list (the same length as 
seeds 
a vector or integers that can be used to set the seed during each search. The number of seeds must be equal to the number of resamples plus one. 
holdout 
the proportion of data in [0, 1) to be heldback from

genParallel 
if a parallel backend is loaded and available, should

allowParallel 
if a parallel backend is loaded and available, should the function use it? 
improve 
the number of iterations without improvement before

An echo of the parameters specified
Max Kuhn
http://topepo.github.io/caret/featureselectionusinggeneticalgorithms.html, http://topepo.github.io/caret/featureselectionusingsimulatedannealing.html
safs
, safs
, , caretGA
,
rfGA
, treebagGA
, caretSA
,
rfSA
, treebagSA
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.