Data Splitting functions
A series of test/training partitions are created using
createResample creates one or
more bootstrap samples.
createFolds splits the data into
createDataPartition(y, times = 1, p = 0.5, list = TRUE, groups = min(5, length(y))) createResample(y, times = 10, list = TRUE) createFolds(y, k = 10, list = TRUE, returnTrain = FALSE) createMultiFolds(y, k = 10, times = 5)
- a vector of outcomes
- the number of partitions to create
- the percentage of data that goes to training
- logical - should the results be in a list (
TRUE) or a matrix with the number of rows equal to
floor(p * length(y))and
- for numeric
y, the number of breaks in the quantiles (see below)
- an integer for the number of folds.
- a logical. When true, the values returned are the
sample positions corresponding to the data used during
training. This argument only works in conjunction with
list = TRUE
For bootstrap samples, simple random sampling is used.
For other data splitting, the random sampling is done within the
y is a factor in an attempt to balance
the class distributions within the splits. For numeric
sample is split into
groups sections based
on quantiles and sampling is done within these subgroups. Also, for
very small class sizes (<= 3)="" the="" classes="" may="" not="" show="" up="" in="" both="" training="" and="" test="" data<="" p="">
For multiple k-fold cross-validation, completely independent folds are created.
The names of the list objects will denote the fold membership using the pattern
"Foldi.Repj" meaning the ith section (of k) of the jth cross-validation set
times). Note that this function calls
list = TRUE and
returnTrain = TRUE.
- A list or matrix of row position integers corresponding to the training data
data(oil) createDataPartition(oilType, 2) x <- rgamma(50, 3, .5) inA <- createDataPartition(x, list = FALSE) plot(density(x[inA])) rug(x[inA]) points(density(x[-inA]), type = "l", col = 4) rug(x[-inA], col = 4) createResample(oilType, 2) createFolds(oilType, 10) createFolds(oilType, 5, FALSE) createFolds(rnorm(21))