Return the value of the label assigned to the input data.

Method call format


Type of return value

A one-dimensional array. The type of data in the array depends on the machine learning task being solved:
  • Regression and ranking  — Numeric values.
  • Binary classification — Numeric values.

    The interpretation of numeric values depends on the selected loss function:

    • Logloss — The value is considered a positive class if it is strictly grater than the value of the border parameter of the loss function. Otherwise, it is considered a negative class.
    • CrossEntropy — The value is interpreted as the probability that the dataset object belongs to the positive class. Possible values are in the range [0; 1].
  • Multiclassification — Integers or strings that represents the labels of the classes.