Generic cross-validation function
cv(
models,
data,
response = NULL,
nfolds = 5,
rep = 1,
weights = NULL,
model.score = NULL,
seed = NULL,
shared = NULL,
args.pred = NULL,
args.future = list(),
mc.cores,
...
)List of fitting functions
data.frame or matrix
Response variable (vector or name of column in data).
Number of folds (default 5. K=0 splits in 1:n/2, n/2:n with last part used for testing)
Number of repetitions (default 1)
Optional frequency weights
Model scoring metric (default: MSE / Brier score). Must be a function with arguments: response, prediction, weights, object, ...
Random seed (argument parsed to future_Apply::future_lapply)
Function applied to each fold with results send to each model
Optional arguments to prediction function (see details below)
Arguments to future.apply::future_mapply
Optional number of cores. parallel::mcmapply used instead of future
Additional arguments parsed to models in models
An object of class 'cross_validated' is returned. See
cross_validated-class for more details about this class and
its generic functions.
models should be list of objects of class ml_model. Alternatively, each element of models should be a list with a fitting function and a prediction function.
The response argument can optionally be a named list where the name is
then used as the name of the response argument in models. Similarly, if data
is a named list with a single data.frame/matrix then this name will be used
as the name of the data/design matrix argument in models.
f0 <- function(data,...) lm(...,data=data)
f1 <- function(data,...) lm(Sepal.Length~Species,data=data)
f2 <- function(data,...) lm(Sepal.Length~Species+Petal.Length,data=data)
x <- cv(list(m0=f0,m1=f1,m2=f2),rep=10, data=iris, formula=Sepal.Length~.)
x
#> Call: cv(models = list(m0 = f0, m1 = f1, m2 = f2), data = iris, rep = 10,
#> formula = Sepal.Length ~ .)
#>
#> 5-fold cross-validation with 10 repetitions
#>
#> mse mae
#> m0 0.09955688 0.2553216
#> m1 0.27017313 0.4062745
#> m2 0.11754607 0.2769748