validate | R Documentation |
validate
is a generic function for cross-validating predictions from
the results of various model fitting functions. The function invokes
particular methods
which depend on the
class
of the first argument.
validate(object, ...)
## S3 method for class 'lm'
validate(
object,
data = NULL,
n_folds = 10,
n_reps = 10,
seed = 42,
silent = FALSE,
...
)
## S3 method for class 'glm'
validate(
object,
data = NULL,
n_folds = 10,
n_reps = 10,
seed = 42,
silent = FALSE,
...
)
## S3 method for class 'zeroinfl'
validate(object, n_folds = 10, n_reps = 10, seed = 42, silent = FALSE, ...)
## S3 method for class 'glmnet'
validate(
object,
x = NULL,
y = NULL,
lambda = NULL,
offset = NULL,
weights = NULL,
n_folds = 10,
n_reps = 10,
seed = 42,
envir = .GlobalEnv,
silent = FALSE,
...
)
## S3 method for class 'beset'
validate(object, ...)
## S3 method for class 'nested'
validate(object, metric = "auto", oneSE = TRUE, ...)
## S3 method for class 'randomForest'
validate(
object,
data = NULL,
x = NULL,
n_folds = 10,
n_reps = 10,
seed = 42,
...,
parallel_type = NULL,
n_cores = NULL,
cl = NULL,
silent = FALSE
)
object |
A model object for which a cross-validated R-squared is desired. |
... |
Additional arguments passed to model summary methods. |
data |
Data frame that was used to train the model. Only needed if
the training data is not contained in the model |
n_folds |
|
n_reps |
|
seed |
|
silent |
object call |
x |
Model matrix that was used to train elastic net. |
y |
Response variable that was used to train elastic net. |
lambda |
|
offset |
|
weights |
= |
envir |
|
metric |
|
oneSE |
|
parallel_type |
(Optional) character string indicating the type of
parallel operation to be used, either |
n_cores |
Integer value indicating the number of workers to run in
parallel during subset search and cross-validation. By default, this will
be set to one fewer than the maximum number of physical cores you have
available, as indicated by |
cl |
(Optional) |
To obtain cross-validation statistics, first fit a model as you
normally would and then pass the model object to validate
.
a "cross_valid
" object consisting of a list with the
following elements:
a list of cross-validated prediction metrics, each containing the mean, between-fold standard error ("btwn_fold_se"), and between-repetition min-max range ("btwn_rep_range") of each metric.
a data frame containing the hold-out predictions for each row in the training data, with a separate column for each repetition of the k-fold cross-validation
a data frame of equal
dimensions to predictions
giving the number of the hold-out fold
of the corresponding element in predictions
a list documenting how many folds and repetitions were used for cross-validation, and the seed passed to the random number generator, which will be needed to reproduce the random fold assignments
validate(lm)
: Cross-validation of linear models
validate(glm)
: Cross-validation of GLMs
validate(zeroinfl)
: Cross-validation of GLMs
validate(glmnet)
: Cross-validation of GLM nets
validate(beset)
: Cross-validation of beset objects
validate(nested)
: Extract test error estimates from "nested beset"
objects with nested cross-validation
validate(randomForest)
: Cross-validation of random forests
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.