Load a tree ensemble model from XGBoost model, stored using the legacy binary format. Note: new XGBoost models should be stored in the JS

Number of features used in the model

Number of decision trees in the model

Output type

Predict with a Treelite model using the General Tree Inference Library (GTIL). Parameters ---------- model : :py:class:`Model` objec

Predict with a Treelite model, outputting the leaf node's ID for each row. Parameters ---------- model : :py:class:`Model` object

Predict with a Treelite model and output prediction of each tree. This function computes one or more margin scores per tree. Parameters

Set a field Parameters ---------- name: Name of the field. Consult :doc:`the model spec </serialization/

Show file download progress.

Returns a strategy to generate datasets for multi-target binary classification, where each target is associated with a binary class label.

Default hypothesis settings

Specify the weighted sample count or the sum of Hessians for the data points that are mapped to the current node. Parameters

Treelite should throw an exception when XGBoost model uses categorical encoding

Test if we can handle extra fields in XGBoost JSON model file Insert an extra field at a random place and then load the model into Treelite,

Test getter and setter methods

Test for invalid node IDs

Test a small random forest with leaf vector output

Test LightGBM binary classifier

Test LightGBM with toy categorical data

Test LightGBM model with depth 32+

Test LightGBM multi-class classifier

Test LightGBM regressor

Test LightGBM with high-cardinality categorical features

Generate a LightGBM ranking model with highly sparse data.

Ensure that Treelite can load old XGBoost models

Test model concatenation with a tree stump

Test model concatenation with a tree stump

Test for orphaned nodes

Test predict_leaf / predict_per_tree with XGBoost binary classifier

Test predict_leaf / predict_per_tree with XGBoost multiclass classifier (grove-per-class)

Test predict_leaf / predict_per_tree with XGBoost regressor

Test whether Treelite objects can be serialized to a buffer

Test whether Treelite objects can be serialized to a byte sequence

Test whether Treelite objects can be serialized to a checkpoint

Scikit-learn classifier

Scikit-learn isolation forest

Scikit-learn isolation forest with feature subsampling

Round trip with scikit-learn RF classifier

Round trip with scikit-learn RF classifier, with multiple outputs and classes

Round trip with scikit-learn RF regressor

Scikit-learn HistGradientBoostingClassifier, with categorical splits

Scikit-learn HistGradientBoostingClassifier, with string categorical features

Scikit-learn RF classifier with multiple outputs and classes

Scikit-learn regressor

Test ability to edit trees

Test XGBoost DART model with dummy data

Test functionality to detect the format of XGBoost (JSON vs. UBJSON)

Test XGBoost with multi-target binary classification problem

Test XGBoost with multi-target regression problem

Test XGBoost with multi-class classification problem

Test XGBoost with non-linear objectives with synthetic data

Test XGBoost with regression data

Throw an error/warning if there are positional arguments after the asterisk. Parameters ---------- func : functio