* Log a CB example asynchronously * @param example - CB example object

Set the simple label. @param weight weight of this example. @param label value of the label (e.g. -1, 1 for binary).

If this example hasn't already been setup (ie, quadratic features constructed, etc.), do so.

This function can take the output of flatten_namespace_extents and reverse it to get back to the original representation required in the features obje

* @brief Wrap text by whole words with the given column width. * * @param text text to wrap * @param width column width to wrap to * @param wrap_a

<summary> Signal that no more examples are send. </summary> <returns>Task completes once the learning and cleanup is done.</returns>

<summary> Performs application-defined tasks associated with freeing, releasing, or resetting unmanaged resources. </summary>

<summary> Returns the number of action dependent examples found within <paramref name="json"/>. </summary> <param name="json">The JSON to be inspected

<summary> Gets an already allocated instance from the example pool or creates a new one. </summary> <returns></returns>

<summary> Marshals a boolean value into native VW. e.g. loggedIn = true yields "loggedIn" in VW native string format. e.g. loggedIn = false yields an

<summary> </summary> <param name="context">The marshalling context.</param> <param name="ns">The namespace description.</param> <param name="featureV

<summary> Predicts for the given example. </summary> <param name="example">The example to predict for.</param> <param name="label">This label can be u

Have to marshal bools, C++ considers them 4 byte quantities, and C# considers them 1 byte.

VW model constructor, exposing all supported parameters to keep sklearn happy Args: convert_to_vw (bool): flag to convert X input

for meta-tasks:

Binary search to find the largest weight w such that w * (fhat^2 - (fhat - w * sens)^2) <= delta. See Section 7.1 in https://arxiv.org/pdf/1703.01014.

<p> Close the underlying VW model. </p> <p> May throw an unchecked exception if the underlying native code throws an Exception. </p> @return <code>tr

Signals the end of the current pass over the data.

Execute a test case and all its dependencies.

Fit the model according to the given training data. Args: X : {array-like, sparse matrix} of shape (n_samples, n_feature