* Calls VW learn on the example and updates the model * @param example - Parsed example from parse()

notify VW that you are done with the example.

* Calls VW predict on the example * @param example - Parsed example from parse()

<summary> Adds an object to the end of the list. </summary> <param name="item">The object to be added to the list.</param> <remarks>Throws an <see cre

get() is only needed for sparse_weights, same as operator[] for lazy_gaussian

a unified function for read(in binary), write(in binary), and write(in text)

This will close the underlying VW model after performing any Close should be written in terms of closeAsync, followed by blocking for the result.

Perform an online update Args: ec: Examples on which the model gets updated. If using a single object the learner must be a singl

* @brief Get the next random value in the range [0,1] and update the PRG state * * @return float random number in range [0,1] */

First create the hash of a namespace.

============================================================ Helper: learn a simple sequence example group ===========================================

Note: does not work for vector options.

Learn from the example then return the prediction given the example, after the internal learner is updated. @param example an example from which to le

============================================================ Helper: learn+finish a multi_ex for cb_explore_adf tests ================================

* Deletes the underlying VW instance. Must be called when done.

This function will return the number of input files AS WELL AS the number of output files. (because of legacy)