A regressor that estimates the target by learning rules to split the data.
SDKs
- macOS 10.14+
- Xcode 10.0+
Framework
- Create ML
Declaration
struct MLDecisionTreeRegressor
StructureMLDecision
MLDecision
Topics
Creating and Training a Decision Tree Regressor
init(trainingCreates a Decision Tree Regressor from the feature columns in the training data to predict the values in the target column.
struct MLDecisionParameters that affect the process of training a model.
let modelThe underlying parameters used when training the model.
var targetThe name of the column you selected at initialization to define which feature the regressor predicts.
var featureThe names of the columns you selected at initialization to train the regressor.
Assessing Model Accuracy
var trainingMeasurements of the regressor’s performance on the training data set.
var validationMeasurements of the regressor’s performance on the validation data set.
Evaluating a Decision Tree Regressor
func evaluation(on: MLDataReturns metrics describing the regressor’s performance on the provided labeled data.
Testing a Decision Tree Regressor
func predictions(from: MLDataPredicts the target value from the provided data.
Saving a Decision Tree Regressor
func write(to: URL, metadata: MLModelExports a Core ML model file for use in your app.
func write(toExports a Core ML model file for use in your app.
Describing a Decision Tree Regressor
var model: MLModelThe underlying Core ML model stored in memory.
var description: StringA text representation of the decision tree regressor.
var debugA text representation of the decision tree regressor that’s suitable for output during debugging.
var playgroundA description of the decision tree regressor shown in a playground.
Relationships
See Also
Supporting Regressor Types
struct MLLinearA regressor that estimates the target as a linear function of the features.
struct MLRandomA regressor based on a collection of decision trees trained on subsets of the data.
struct MLBoostedA regressor based on a collection of decision trees combined with gradient boosting.