Oppublic final class BoostedTreesUpdateEnsemble extends PrimitiveOp
or by starting a new tree.
operation| Modifier and Type | Method | Description |
|---|---|---|
static BoostedTreesUpdateEnsemble |
create(Scope scope,
Operand<?> treeEnsembleHandle,
Operand<java.lang.Integer> featureIds,
java.lang.Iterable<Operand<java.lang.Integer>> nodeIds,
java.lang.Iterable<Operand<java.lang.Float>> gains,
java.lang.Iterable<Operand<java.lang.Integer>> thresholds,
java.lang.Iterable<Operand<java.lang.Float>> leftNodeContribs,
java.lang.Iterable<Operand<java.lang.Float>> rightNodeContribs,
Operand<java.lang.Integer> maxDepth,
Operand<java.lang.Float> learningRate,
java.lang.Long pruningMode) |
Factory method to create a class to wrap a new BoostedTreesUpdateEnsemble operation to the graph.
|
clone, finalize, getClass, notify, notifyAll, wait, wait, waitequals, hashCode, toStringpublic static BoostedTreesUpdateEnsemble create(Scope scope, Operand<?> treeEnsembleHandle, Operand<java.lang.Integer> featureIds, java.lang.Iterable<Operand<java.lang.Integer>> nodeIds, java.lang.Iterable<Operand<java.lang.Float>> gains, java.lang.Iterable<Operand<java.lang.Integer>> thresholds, java.lang.Iterable<Operand<java.lang.Float>> leftNodeContribs, java.lang.Iterable<Operand<java.lang.Float>> rightNodeContribs, Operand<java.lang.Integer> maxDepth, Operand<java.lang.Float> learningRate, java.lang.Long pruningMode)
scope - current graph 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.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.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.