Test the categories container for iterator-based inputs.

Run checks with mixed types.

Basic tests for invalid inputs.

Basic tests for leaf prediction.

Test OOV categories whose sorted insertion point is inside the training set.

Basic tests for re-coding during prediction.

Basic tests for SHAP values.

Basic tests for thread safety.

Tests plotting functions for categorical features.

Test column sampling with feature importance for multi-target trees.

Tests custom objective and metric functions.

Test for learning rate.

Test learning rate scheduler, used by both CPU and GPU tests.

check decay has effect on leaf output.

Exp family has a closed solution.

Check that reparameterized expectile regression produces non-crossing curves.

Different strategies produce similar feature importance ratios.

Run federated learning tests.

Test that gradient-based sampling provides better accuracy.

Test grow policy (depthwise and lossguide) for vector leaf.

Testwith the cali housing dataset.

Test for init estimation.

Tests for the intercept.

Test with invalid categorical inputs.

Test https://github.com/dmlc/xgboost/issues/11499 .

Test data with size smaller than number of categories.

Check QDM with mixed batches.

Test mixed multi_strategy with ResetStrategy callback.

Use vector leaf for multi-class models.

Use vector leaf for multi-label classification models.

Test normalization.

Test for PR AUC metric on a binary classification problem.

Test for PR AUC metric on a ranking problem.

Test for PR AUC metric on a multi-class classification problem.

Check quantile regression for vector leaf.

Test LTR with categorical features.

Test ranking with qid packed into X.

Run re-coding test with the Dask interface.

Test re-coding inpput for DMatrix.

Run prediction with re-coded DMatrix.

Test re-coding for training continuation.

Basic test for using reduced gradient for tree splits.

TestROC AUC metric on a binary classification problem.

Test for ROC AUC metric on a multi-class classification problem.

Test normalization by score differences.

Run with manually specified category encoding.

Tests split_value_histograms with categorical features.

Test row subsampling.

Test re-coding for training continuation.

Check that 2-session training (4+4 iters) equals single-session (8 iters).

Run training continuation test.

Tests tree_to_df with categorical features.

Test with individual updaters.

Check the validation dataset is using the correct encoding.

Test vector leaf with external memory.

Save the metadata of an xgboost.spark._SparkXGBEstimator or xgboost.spark._SparkXGBModel.

Set parameter to the Booster. @param key param name @param value param value @throws XGBoostError native error

Set parameters to the Booster. @param params parameters key-value map @throws XGBoostError native error

Iterate through folds for setting attributes

Set device, optional value: cpu, cuda, gpu

! * \brief set input stream */

Simulate click data using position biased model (PBM).

Slice the DMatrix and return a new DMatrix that only contains `rowIndex`. @param rowIndex row index @return sliced new DMatrix @throws XGBoostError n

Split value of a node.

Convert specified tree to graphviz instance. IPython can automatically plot the returned graphviz instance. Otherwise, you should call .render() m

Convert a sequence of bytes to a list of Python integer

Train a multi-output interval model.

Test accuracy, reused by GPU tests.

Basic training continuation.

Training continuation with early stopping.

Attempt to fetch JVM docs for a given branch. Returns True if successful