Create DMatrix from spark data partitions. Parameters ---------- iterator : Pyspark partition iterator. feature_cols:

Cross-validation with given parameters. @param data Data to be trained. @param params Booster params. @param round Number of boosting iteration

Convert ctypes pointer to buffer type.

Train using Python implementation of Squared Error.

Cross-validation with given parameters. Parameters ---------- params : dict Booster params. dtrain : Data to be train

Get the data split mode of the DMatrix. .. versionadded:: 2.1.0

Look for the demo directory based on the test file name.

Detects the chip architecture using the system properties. Throws IllegalStateException if the architecture is not recognized. @return The architectur

Detects the OS using the system properties. Throws IllegalStateException if the OS is not recognized. @return The OS.

The total number of bytes of memory this GPU has.

transfer DMatrix array to handle array (used for native functions) @param dmatrixs @return handle array for input dmatrixs

Dump model into a text or JSON file. Unlike :py:meth:`save_model`, the output format is primarily used for visualization or interpretation,

NOLINT

The function to predict.

flatten a mat to array

Generate a sample dataset that y is a 2 dim circle.

Get attributes stored in the Booster as a Map. @return A map contain attribute pairs. @throws XGBoostError native error

Get the importance of each feature based on information gain or cover @return featureImportanceMap key: feature index, value: feature importance scor

Get feature names @return an array of feature names to be returned @throws XGBoostError

Get feature types @return an array of feature types to be returned @throws XGBoostError

user define objective function, return gradient and second order gradient @param predicts untransformed margin predicts @param dtrain training data

Returns a port that's available during the function call. It doesn't prevent the port from being used after the function returns as we can't reser

Guess the git branch.

Generate reproducible synthetic classification data.

Internal method for retrieving the handle.

Get the logger level for the given log name

Get the commit hash for the current documentation build.

Given group row boundaries, convert ground indexes to row indexes :param groups: list of groups for testing :param boundaries: rows index

Initialize the collective communicator on current working thread. @param envs The additional environment variables to pass to the communicator. @thro

Is this an arrow dictionary array?

Whether a node has categorical split.

Check cuDF package available or not

Check cupy package available or not

Check whether dtype is a dictionary type.

Return whether dtype is a pyarrow extension type for pandas

Check pyarrow package available or not

Predicate for scipy COO input.

Predicate for scipy CSC input.

Load the metadata and the instance of an xgboost.spark._SparkXGBEstimator or xgboost.spark._SparkXGBModel. :return: a tuple

Load the MSLR10k dataset from data_path and save parquet files in the cache_path.

Load the MSLR10k dataset from data_path and cache a pickle object in cache_path. Returns ------- A list of tuples [(X, y, qid), ...].

Locate libxgboost from the local project directory's lib/ subdirectory.

Decorate a learning to rank metric.

Entry point for training.

Make an __(cuda)_array_interface__ from a pointer.

Generate categorical features for test. Parameters ---------- n_categories: Number of categories for categorical features. on

Make some random data for demo.

Factory function for creating strategies that generates datasets with weight and base margin.

Generate data for demo.

A function to prevent query group from being scattered to different workers. Please see the tutorial in the document for the implication for not h

Handle pandas pyarrow extention.

Enumerate all supported pandas extension types.

Use the reference categories from the model. If none, then use the reference categories from the training DMatrix. Parameters ----------

Change makefiles according to the package layout.

Function to visualize censored labels

Predict contributions with data with the full model.

predict with the given DMatrix @param testSet the local test set represented as DMatrix @return prediction result

Predict leaf indices given the data @param data The input data. @param treeLimit Number of trees to include, 0 means all trees. @return The leaf indi

Print all recorded items by :py:func:`record_time`.

Process a file.

Retrieve a file from a URL with progress indication.

Draw a circle with 2-dim coordinate as target variables.

Test mean absolute error with vector leaf.

Test for adaptive trees.

Test for the relation between score and margin.

Basic tests for the container class used by the DMatrix.