Create a partial response or effect plot of a trained model.
Usage
partial(
mod,
x.var = NULL,
constant = NULL,
variable_length = 100,
values = NULL,
newdata = NULL,
plot = FALSE,
type = "response",
...
)
# S4 method for class 'ANY'
partial(
mod,
x.var = NULL,
constant = NULL,
variable_length = 100,
values = NULL,
newdata = NULL,
plot = FALSE,
type = "response",
...
)
partial.DistributionModel(mod, ...)
Arguments
- mod
A trained
DistributionModel
object withfit_best
model within.- x.var
A character indicating the variable for which a partial effect is to be calculated.
- constant
A numeric constant to be inserted for all other variables. Default calculates a mean per variable.
- variable_length
numeric The interpolation depth (nr. of points) to be used (Default:
100
).- values
numeric Directly specified values to compute partial effects for. If this parameter is set to anything other than
NULL
, the parameter"variable_length"
is ignored (Default:NULL
).- newdata
An optional data.frame with provided data for partial estimation (Default:
NULL
).- plot
A
logical
indication of whether the result is to be plotted?- type
A specified type, either
'response'
or'predictor'
. Can be missing.- ...
Other engine specific parameters.
Value
A data.frame with the created partial response.