Description
The Forest Analyze function analyzes the model generated by the Forest Drive function and assigns weights to the model variables. The variable/attribute counts in each tree level help you understand the importance of different variables in the decision-making process.
Usage
td_decision_forest_evaluator_mle ( object = NULL, num.levels = 5, object.sequence.column = NULL )
Arguments
object |
Required Argument. |
num.levels |
Optional Argument. |
object.sequence.column |
Optional Argument. |
Value
Function returns an object of class "td_decision_forest_evaluator_mle"
which is a named list containing Teradata tbl object.
Named list member can be referenced directly with the "$" operator
using name: result
Examples
# Get the current context/connection con <- td_get_context()$connection # Load the data to run the example loadExampleData("decisionforest_example","housing_train") # Create remote tibble objects. housing_train <- tbl(con, "housing_train") # First train the data, i.e., create a decision forest model td_decision_forest_out <- td_decision_forest_mle(data=housing_train, formula = (homestyle ~ bedrooms + lotsize + gashw + driveway + stories + recroom + price + garagepl + bathrms + fullbase + airco + prefarea), tree.type="classification", nodesize=1, variance=0, max.depth=12, maxnum.categorical=20, mtry = 100, seed = 100, ntree=50 ) # Analyze generated output. decision_forest_evaluator_out <- td_decision_forest_evaluator_mle(object=td_decision_forest_out, num.levels=5 )