xgboost.testing.updater Namespace Reference

Functions
float	get_basescore (xgb.XGBModel model)
None	check_init_estimation (str tree_method)
None	check_quantile_loss (str tree_method, bool weighted)
None	check_cut (int n_entries, np.ndarray indptr, np.ndarray data, Any dtypes)
None	check_get_quantile_cut_device (str tree_method, bool use_cupy)
None	check_get_quantile_cut (str tree_method)
None	check_categorical_ohe (int rows, int cols, int rounds, int cats, str device, str tree_method)
None	check_categorical_missing (int rows, int cols, int cats, str device, str tree_method)
Dict[str, Any]	train_result (Dict[str, Any] param, xgb.DMatrix dmat, int num_rounds)

Variables
	USE_ONEHOT = np.iinfo(np.int32).max
int	USE_PART = 1

Detailed Description

Tests for updaters.

Function Documentation

check_categorical_missing()

None xgboost.testing.updater.check_categorical_missing	(	int	rows,
		int	cols,
		int	cats,
		str	device,
		str	tree_method
	)

Check categorical data with missing values.

check_cut()

None xgboost.testing.updater.check_cut	(	int	n_entries,
		np.ndarray	indptr,
		np.ndarray	data,
		Any	dtypes
	)

Check the cut values.

check_get_quantile_cut()

None xgboost.testing.updater.check_get_quantile_cut

(

str

tree_method

)

Check the quantile cut getter.

check_get_quantile_cut_device()

None xgboost.testing.updater.check_get_quantile_cut_device	(	str	tree_method,
		bool	use_cupy
	)

Check with optional cupy.

check_init_estimation()

None xgboost.testing.updater.check_init_estimation

(

str

tree_method

)

Test for init estimation.

check_quantile_loss()

None xgboost.testing.updater.check_quantile_loss	(	str	tree_method,
		bool	weighted
	)

Test for quantile loss.

get_basescore()

float xgboost.testing.updater.get_basescore

(

xgb.XGBModel

model

)

Get base score from an XGBoost sklearn estimator.

train_result()

Dict[str, Any] xgboost.testing.updater.train_result	(	Dict[str, Any]	param,
		xgb.DMatrix	dmat,
		int	num_rounds
	)

Get training result from parameters and data.