Which Spark MLlib class implements Gradient Boosted Trees for classification tasks?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Spark MLlib uses the abbreviation GBT for Gradient Boosted Trees — the class is GBTClassifier, found in pyspark.ml.classification, following the same naming convention as DecisionTreeClassifier and RandomForestClassifier.
Full explanation below image
Full Explanation
pyspark.ml.classification.GBTClassifier is the Spark MLlib implementation of Gradient Boosted Trees for binary classification (note: Spark's GBT only supports binary classification, not multiclass). Key hyperparameters: maxIter (number of trees/boosting rounds), maxDepth (tree depth, default 5), stepSize (learning rate, default 0.1), subsamplingRate (fraction of data per iteration). For regression: pyspark.ml.regression.GBTRegressor. For multiclass classification with boosting, use XGBoost via SynapseML or spark-xgboost. GBTClassifier follows the Estimator/Transformer pattern: gbt = GBTClassifier(featuresCol='features', labelCol='label', maxIter=20); model = gbt.fit(train_df); predictions = model.transform(test_df). The other class names are fabricated — GradientBoostingClassifier is scikit-learn's name.