ml_options {sparklyr} | R Documentation |
Provide this object to the various Spark ML methods, to control certain facets of the model outputs produced.
ml_options(id.column = random_string("id"), response.column = random_string("response"), features.column = random_string("features"), model.transform = NULL, only.model = FALSE, na.action = getOption("na.action", "na.omit"), ...)
id.column |
The name to assign to the generated id column. |
response.column |
The name to assign to the generated response column. |
features.column |
The name to assign to the generated features column. |
model.transform |
An optional R function that accepts a Spark model
and returns a Spark model. This can be used to supply optional Spark model
fitting parameters not made available in the |
only.model |
Boolean; should the Spark model object itself be returned
without fitting the actual model? Useful for |
na.action |
An R function, or the name of an R function, indicating how missing values should be handled. |
... |
Optional arguments, reserved for future expansion. |