converters.xgboost._tree_ensemble

Module Contents

Functions

recurse_json(mlkit_tree,xgb_tree_json,tree_id,node_id,feature_map,force_32bit_float) Traverse through the tree and append to the tree spec.
convert_tree_ensemble(model,feature_names,target,force_32bit_float) Convert a generic tree model to the protobuf spec.
recurse_json(mlkit_tree, xgb_tree_json, tree_id, node_id, feature_map, force_32bit_float)

Traverse through the tree and append to the tree spec.

convert_tree_ensemble(model, feature_names, target, force_32bit_float)

Convert a generic tree model to the protobuf spec.

This currently supports:
  • Decision tree regression
model: str | Booster
Path on disk where the XGboost JSON representation of the model is or a handle to the XGboost model.
feature_names : list of strings or None
Names of each of the features. When set to None, the feature names are extracted from the model.
target: str,
Name of the output column.
force_32bit_float: bool
If True, then the resulting CoreML model will use 32 bit floats internally.
model_spec: An object of type Model_pb.
Protobuf representation of the model