! * \brief get un-transformed prediction * \param data training data matrix * \param out_preds output vector that stores the prediction *

! * \brief Add sorted column entries into this sketch. * * \param column Sorted column entries in ascending order by feature value. * \par

! \brief push up a prepared summary */

* @brief Obtain a view to the sorted index created by the @ref Sort method. */

Resize without keeping the data*/

! * \brief combine `other` into `this`. * * \param other Input summary to combine with `this`. * \param workspace Optional entry buffer fo

! * \brief Rebind underlying storage span while preserving current logical size. */

For testing, getter for the number of fetches for sparse page source.

!\return feature index to split on */

* @brief Get a thread local buffer. * * @param n The size of the thread local block. */

* @brief CPU implementation of update prediction cache, which calculates the leaf value * for the last tree and accumulates it to prediction v

Sanity checks

The following two functions are `Load` and `Save` for memory based serialization methods. E.g. Python pickle.

Create from data iterator

A switch function for async environment.

Validate parameters in distributed environments.