ml_tree_tidiers

Tidying methods for Spark ML tree models

Description

These methods summarize the results of Spark ML models into tidy forms.

Usage

tidyml_model_decision_tree_classification(x, …)

tidyml_model_decision_tree_regression(x, …)

augmentml_model_decision_tree_classification(x, newdata = NULL, …)

augmentml_model_decision_tree_regression(x, newdata = NULL, …)

glanceml_model_decision_tree_classification(x, …)

glanceml_model_decision_tree_regression(x, …)

tidyml_model_random_forest_classification(x, …)

tidyml_model_random_forest_regression(x, …)

augmentml_model_random_forest_classification(x, newdata = NULL, …)

augmentml_model_random_forest_regression(x, newdata = NULL, …)

glanceml_model_random_forest_classification(x, …)

glanceml_model_random_forest_regression(x, …)

tidyml_model_gbt_classification(x, …)

tidyml_model_gbt_regression(x, …)

augmentml_model_gbt_classification(x, newdata = NULL, …)

augmentml_model_gbt_regression(x, newdata = NULL, …)

glanceml_model_gbt_classification(x, …)

glanceml_model_gbt_regression(x, …)

Arguments

Argument Description
x a Spark ML model.
extra arguments (not used.)
newdata a tbl_spark of new data to use for prediction.