public final class BoostedTreesUpdateEnsembleV2 extends PrimitiveOp
or by starting a new tree.
operation| Modifier and Type | Method and Description |
|---|---|
static BoostedTreesUpdateEnsembleV2 |
create(Scope scope,
Operand<?> treeEnsembleHandle,
Operand<Integer> featureIds,
Iterable<Operand<Integer>> dimensionIds,
Iterable<Operand<Integer>> nodeIds,
Iterable<Operand<Float>> gains,
Iterable<Operand<Integer>> thresholds,
Iterable<Operand<Float>> leftNodeContribs,
Iterable<Operand<Float>> rightNodeContribs,
Iterable<Operand<String>> splitTypes,
Operand<Integer> maxDepth,
Operand<Float> learningRate,
Operand<Integer> pruningMode)
Factory method to create a class wrapping a new BoostedTreesUpdateEnsembleV2 operation.
|
equals, hashCode, op, toStringpublic static BoostedTreesUpdateEnsembleV2 create(Scope scope, Operand<?> treeEnsembleHandle, Operand<Integer> featureIds, Iterable<Operand<Integer>> dimensionIds, Iterable<Operand<Integer>> nodeIds, Iterable<Operand<Float>> gains, Iterable<Operand<Integer>> thresholds, Iterable<Operand<Float>> leftNodeContribs, Iterable<Operand<Float>> rightNodeContribs, Iterable<Operand<String>> splitTypes, Operand<Integer> maxDepth, Operand<Float> learningRate, Operand<Integer> pruningMode)
scope - current scopetreeEnsembleHandle - Handle to the ensemble variable.featureIds - Rank 1 tensor with ids for each feature. This is the real id of
the feature that will be used in the split.dimensionIds - List of rank 1 tensors representing the dimension in each feature.nodeIds - List of rank 1 tensors representing the nodes for which this feature
has a split.gains - List of rank 1 tensors representing the gains for each of the feature's
split.thresholds - List of rank 1 tensors representing the thesholds for each of the
feature's split.leftNodeContribs - List of rank 2 tensors with left leaf contribs for each of
the feature's splits. Will be added to the previous node values to constitute
the values of the left nodes.rightNodeContribs - List of rank 2 tensors with right leaf contribs for each
of the feature's splits. Will be added to the previous node values to constitute
the values of the right nodes.splitTypes - List of rank 1 tensors representing the split type for each feature.maxDepth - Max depth of the tree to build.learningRate - shrinkage const for each new tree.pruningMode - 0-No pruning, 1-Pre-pruning, 2-Post-pruning.Copyright © 2015–2019. All rights reserved.