API Reference
Documentation for the OptimalTrees public interface.
Index
IAI.FeatureSetIAI.OptimalTreeClassifierIAI.OptimalTreeLearnerIAI.OptimalTreePolicyMaximizerIAI.OptimalTreePolicyMinimizerIAI.OptimalTreePrescriptionMaximizerIAI.OptimalTreePrescriptionMinimizerIAI.OptimalTreeRegressorIAI.OptimalTreeSurvivalLearner
Learners
IAI.FeatureSet — TypePermissible types for specifying set of features in a dataframe. Refer to Indexing in DataFrames.jl for a full list of supported rules:
| Input Type | Description | Examples |
|---|---|---|
All | Use all columns | All() |
Integer or a vector of Integers | Specify indices of columns to use | 1, [1, 3, 4] |
Symbol or a vector of Symbols | Specify names of columns to use | :x1, [:x1, :x3] |
String or a vector of Strings | Specify names of columns to use | "x1", ["x1", "x3"] |
Not | Specify columns not to use | Not(1), Not(["x2", "x4"]) |
Between | Specify range of columns to use | Between("x1", "x4") |
IAI.OptimalTreeLearner — TypeAbstract type encompassing all optimal tree learners.
IAI.OptimalTreeClassifier — TypeLearner for training Optimal Classification Trees.
The following parameters are supported (refer to the documentation for each):
IAI.OptimalTreeRegressor — TypeLearner for training Optimal Regression Trees.
The following parameters are supported (refer to the documentation for each):
IAI.OptimalTreeSurvivalLearner — TypeLearner for training Optimal Survival Trees.
The following parameters are supported (refer to the documentation for each):
IAI.OptimalTreePrescriptionMinimizer — TypeLearner for training Optimal Prescriptive Trees where the prescriptions should aim to minimize outcomes.
The following parameters are supported (refer to the documentation for each):
IAI.OptimalTreePrescriptionMaximizer — TypeLearner for training Optimal Prescriptive Trees where the prescriptions should aim to maximize outcomes.
Supports the same parameters as OptimalTreePrescriptionMinimizer.
IAI.OptimalTreePolicyMinimizer — TypeLearner for training Optimal Policy Trees where the prescriptions should aim to minimize outcomes.
The following parameters are supported (refer to the documentation for each):
IAI.OptimalTreePolicyMaximizer — TypeLearner for training Optimal Policy Trees where the prescriptions should aim to maximize outcomes.
Supports the same parameters as OptimalTreePolicyMinimizer.