package tensorflow.boosted_trees.trees

Get desktop application:
View/edit binary Protocol Buffers messages

Split rule for categorical features with a single feature Id.

Used in: TreeNode

• int32 feature_column = 1

  Categorical feature column and Id describing the rule feature == Id.

• int64 feature_id = 2

• int32 left_id = 3

  Node children indexing into a contiguous vector of nodes starting from the root.

• int32 right_id = 4

Split rule for categorical features with a set of feature Ids.

Used in: TreeNode

• int32 feature_column = 1

  Categorical feature column and Id describing the rule feature ∈ feature_ids.

• repeated int64 feature_ids = 2

  Sorted list of Ids in the set.

• int32 left_id = 3

  Node children indexing into a contiguous vector of nodes starting from the root.

• int32 right_id = 4

DecisionTreeConfig describes a list of connected nodes. Node 0 must be the root and can carry any payload including a leaf in the case of representing the bias. Note that each node id is implicitly its index in the list of nodes.

Used in: DecisionTreeEnsembleConfig

• repeated TreeNode nodes = 1

DecisionTreeEnsembleConfig describes an ensemble of decision trees.

• repeated DecisionTreeConfig trees = 1

• repeated float tree_weights = 2

• repeated DecisionTreeMetadata tree_metadata = 3

• optional GrowingMetadata growing_metadata = 4

  Metadata that is used during the training.

Used in: DecisionTreeEnsembleConfig

• int32 num_tree_weight_updates = 1

  How many times tree weight was updated (due to reweighting of the final ensemble, dropout, shrinkage etc).

• int32 num_layers_grown = 2

  Number of layers grown for this tree.

• bool is_finalized = 3

  Whether the tree is finalized in that no more layers can be grown.

Split rule for dense float features.

Used in: SparseFloatBinarySplitDefaultLeft, SparseFloatBinarySplitDefaultRight, TreeNode

• int32 feature_column = 1

  Float feature column and split threshold describing the rule feature <= threshold.

• float threshold = 2

• int32 left_id = 3

  Node children indexing into a contiguous vector of nodes starting from the root.

• int32 right_id = 4

Used in: DecisionTreeEnsembleConfig

• int64 num_trees_attempted = 1

  Number of trees that we have attempted to build. After pruning, these trees might have been removed.

• int64 num_layers_attempted = 2

  Number of layers that we have attempted to build. After pruning, these layers might have been removed.

Leaves can either hold dense or sparse information.

Used in: learner.SplitInfo, TreeNode, TreeNodeMetadata

• oneof leaf

  • Vector vector = 1

    See third_party/tensorflow/contrib/decision_trees/ proto/generic_tree_model.proto for a description of how vector and sparse_vector might be used.

  • SparseVector sparse_vector = 2

Split rule for sparse float features defaulting left for missing features.

Used in: TreeNode

• optional DenseFloatBinarySplit split = 1

Split rule for sparse float features defaulting right for missing features.

Used in: TreeNode

• optional DenseFloatBinarySplit split = 1

Used in: Leaf

• repeated int32 index = 1

• repeated float value = 2

TreeNode describes a node in a tree.

Used in: learner.SplitInfo, DecisionTreeConfig

• oneof node

  • Leaf leaf = 1

  • DenseFloatBinarySplit dense_float_binary_split = 2

  • SparseFloatBinarySplitDefaultLeft sparse_float_binary_split_default_left = 3

  • SparseFloatBinarySplitDefaultRight sparse_float_binary_split_default_right = 4

  • CategoricalIdBinarySplit categorical_id_binary_split = 5

  • CategoricalIdSetMembershipBinarySplit categorical_id_set_membership_binary_split = 6

• optional TreeNodeMetadata node_metadata = 777

TreeNodeMetadata encodes metadata associated with each node in a tree.

Used in: TreeNode

• float gain = 1

  The gain associated with this node.

• optional Leaf original_leaf = 2

  The original leaf node before this node was split.

Used in: Leaf

• repeated float value = 1