@@ -230,10 +230,10 @@ object DecisionTree extends Serializable with Logging {
230230 * @return a DecisionTreeModel that can be used for prediction
231231 */
232232 def train (
233- input : RDD [LabeledPoint ],
234- algo : Algo ,
235- impurity : Impurity ,
236- maxDepth : Int ): DecisionTreeModel = {
233+ input : RDD [LabeledPoint ],
234+ algo : Algo ,
235+ impurity : Impurity ,
236+ maxDepth : Int ): DecisionTreeModel = {
237237 val strategy = new Strategy (algo, impurity, maxDepth)
238238 // Converting from standard instance format to weighted input format for tree training
239239 val weightedInput = input.map(x => WeightedLabeledPoint (x.label, x.features))
@@ -279,10 +279,10 @@ object DecisionTree extends Serializable with Logging {
279279 * @param impurity impurity criterion used for information gain calculation
280280 * @param maxDepth maxDepth maximum depth of the tree
281281 * @param numClassesForClassification number of classes for classification. Default value of 2.
282- * @param labelWeights A map storing weights applied to each label for handling unbalanced
282+ * @param labelWeights A map storing weights for each label to handle unbalanced
283283 * datasets. For example, an entry (n -> k) implies the a weight of k is
284284 * applied to an instance with label n. It's important to note that labels
285- * are zero-index and take values 0, 1, 2, ... , numClasses.
285+ * are zero-index and take values 0, 1, 2, ... , numClasses - 1 .
286286 * @return a DecisionTreeModel that can be used for prediction
287287 */
288288 def train (
@@ -316,7 +316,7 @@ object DecisionTree extends Serializable with Logging {
316316 * @param labelWeights A map storing weights applied to each label for handling unbalanced
317317 * datasets. For example, an entry (n -> k) implies the a weight of k is
318318 * applied to an instance with label n. It's important to note that labels
319- * are zero-index and take values 0, 1, 2, ... , numClasses.
319+ * are zero-index and take values 0, 1, 2, ... , numClasses - 1 .
320320 * @param maxBins maximum number of bins used for splitting features
321321 * @param quantileCalculationStrategy algorithm for calculating quantiles
322322 * @param categoricalFeaturesInfo A map storing information about the categorical variables and
0 commit comments