BoostedTreesExampleDebugOutputs

public final class BoostedTreesExampleDebugOutputs

Debugging/model interpretability outputs for each example.

It traverses all the trees and computes debug metrics for individual examples, such as getting split feature ids and logits after each split along the decision path used to compute directional feature contributions.

Constants

String OP_NAME The name of this op, as known by TensorFlow core engine

Public Methods

Output<TString>
asOutput()
Returns the symbolic handle of the tensor.
static BoostedTreesExampleDebugOutputs
create(Scope scope, Operand<?> treeEnsembleHandle, Iterable<Operand<TInt32>> bucketizedFeatures, Long logitsDimension)
Factory method to create a class wrapping a new BoostedTreesExampleDebugOutputs operation.
Output<TString>
examplesDebugOutputsSerialized()
Output rank 1 Tensor containing a proto serialized as a string for each example.

Inherited Methods

Constants

public static final String OP_NAME

The name of this op, as known by TensorFlow core engine

Constant Value: "BoostedTreesExampleDebugOutputs"

Public Methods

public Output<TString> asOutput ()

Returns the symbolic handle of the 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 static BoostedTreesExampleDebugOutputs create (Scope scope, Operand<?> treeEnsembleHandle, Iterable<Operand<TInt32>> bucketizedFeatures, Long logitsDimension)

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

Parameters
scope current scope
bucketizedFeatures A list of rank 1 Tensors containing bucket id for each feature.
logitsDimension scalar, dimension of the logits, to be used for constructing the protos in examples_debug_outputs_serialized.
Returns
  • a new instance of BoostedTreesExampleDebugOutputs

public Output<TString> examplesDebugOutputsSerialized ()

Output rank 1 Tensor containing a proto serialized as a string for each example.