tfdf.py_tree.value.RegressionValue

The regression value of a regressive tree.

Inherits From: AbstractValue

tfdf.py_tree.value.RegressionValue(
    value: float,
    num_examples: Optional[float] = 1.0,
    standard_deviation: Optional[float] = None
)

Can also be used in gradient-boosted-trees for classification and ranking.

value Value of the tree. The semantic depends on the tree: For Random Forests, this value is a regressive value (in the same unit as the label). For classification and ranking GBDTs, this value is a loggit.
standard_deviation Optional standard deviation attached to the value.
num_examples Number of example in the node.

Methods

__eq__

View source

__eq__(
    other
)

Return self==value.