BoostedTreesCenterBias

public final class BoostedTreesCenterBias

Calculates the prior from the training data (the bias) and fills in the first node with the logits' prior. Returns a boolean indicating whether to continue centering.

Public Methods

Output <Boolean>
asOutput ()
Returns the symbolic handle of a tensor.
Output <Boolean>
continueCentering ()
Bool, whether to continue bias centering.
static BoostedTreesCenterBias
create ( Scope scope, Operand <?> treeEnsembleHandle, Operand <Float> meanGradients, Operand <Float> meanHessians, Operand <Float> l1, Operand <Float> l2)
Factory method to create a class wrapping a new BoostedTreesCenterBias operation.

Inherited Methods

Public Methods

public Output <Boolean> asOutput ()

Returns the symbolic handle of a tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

public Output <Boolean> continueCentering ()

Bool, whether to continue bias centering.

public static BoostedTreesCenterBias create ( Scope scope, Operand <?> treeEnsembleHandle, Operand <Float> meanGradients, Operand <Float> meanHessians, Operand <Float> l1, Operand <Float> l2)

Factory method to create a class wrapping a new BoostedTreesCenterBias operation.

Parameters
scope current scope
treeEnsembleHandle Handle to the tree ensemble.
meanGradients A tensor with shape=[logits_dimension] with mean of gradients for a first node.
meanHessians A tensor with shape=[logits_dimension] mean of hessians for a first node.
l1 l1 regularization factor on leaf weights, per instance based.
l2 l2 regularization factor on leaf weights, per instance based.
Returns
  • a new instance of BoostedTreesCenterBias