Returns a snapshot of the current model. @return serialized VW model.

@return the weightedExampleSum

Get driver output messages (training progress, statistics, final results). One line per item in the list. To get the complete log including ru

see gd.cc:audit_feature(..)

The inline function below may be adjusted to change the way synthetic (interaction) features' values are calculated, e.g., fabs(value1-value2) or even

Updates the associated VW model using this example.

<code>learnOrPredict</code> allows the ability to return an unboxed prediction. This will reduce the overhead of this function call. @param example a

<code>learnOrPredict</code> allows the ability to return an unboxed prediction. This will reduce the overhead of this function call. @param example a

Extract argument value from VW command string.

This function is used to move the weights of a specific offset in the innermost feature_width to another offset.

Return the total number of features in this namespace.

Get the total number of features in a given namespace Args: ns : namespace Get the total features of this namespa

https://arxiv.org/pdf/1611.07612.pdf

* Stores the VW model in binary format to a file (Node.js only) * @param model_file - Path to save the model

* @brief Override the current PRG state value * * @param new_state new state value */

* Start logging to a file (appends) * @param log_file - Path to log file

* \brief Check if a string starts with some other string. * \param full_string String to check starting of * \param starting Starting value to check

Convert array or sparse matrix to Vowpal Wabbit format Args: x : {array-like, sparse matrix}, shape (n_samples, n_features)

train the node with id cn, using the statistics stored in the node to formulate a binary classificaiton example.

TODO: ideally we should enable_if this only on .NET Blittable types */

<summary> Creates a new instance of <see cref="VowpalWabbitAsync{TExample}"/> to feed examples of type <typeparamref name="TExample"/>. </summary> <ty

<summary> Create the invocation expression of a marshalling method. </summary> <param name="methodName">The marshalling method to invoke.</param> <par

<summary> Evaluates <paramref name="learnedAction"/> and <paramref name="numActions"/>x constants policies w.r.t. to <paramref name="label"/>. </summa

<summary> Gets the member info in a sort of type safe manner - it's better than using strings, but some runtime errors are still possbile. </summary>

<summary> Registers a parsing extension. </summary> <param name="extension">The extension to be rgistered.</param>

<summary> Serializes the specifed example to VW native string format. </summary> <typeparam name="TExample">The user example type.</typeparam> <typepa

<summary> Updates the model used for prediction in a thread-safe manner. </summary> <param name="model">The new model to be used.</param>