package yggdrasil_decision_forests.model.isolation_forest.proto

Get desktop application:
View/edit binary Protocol Buffers messages

Header for the isolation forest model.

Next ID: 6

Used in: IsolationForestSerializedModel

• optional int32 num_node_shards = 1

  Number of shards used to store the nodes.

• optional int64 num_trees = 2

  Number of trees.

• optional string node_format = 3

  Container used to store the trees' nodes.

• optional int64 num_examples_per_trees = 4

  Number of examples used to grow each tree.

• oneof anomality_score

  • IsolationForestAnomalityScore isolation_forest = 5

An isolation forest as defined in "Isolation-Based Anomaly Detection" by Liu et al. (2012). In this case, the prediction value is the node depth.

Used in: Header

(message has no fields)

• optional Header header = 1

Training configuration for the Isolation Forest algorithm.

Next ID: 5

• optional decision_tree.proto.DecisionTreeTrainingConfig decision_tree = 1

  Decision tree specific parameters.

• optional int32 num_trees = 2

  Number of trees in the forest.

• oneof sampling_method

  Number of examples used to grow each tree. Only one of "subsample_ratio" and "subsample_count" can be set. If neither is set, "subsample_count" is assumed to be equal to 256. This is the default value recommended in the isolation forest paper (https://ieeexplore.ieee.org/abstract/document/4781136).

  • float subsample_ratio = 3

  • int32 subsample_count = 4