Class MultiClassLearner

Primary class for running classification models. It is designed to abstract the separation of training and test sets as well as select best result across all classes.

Inheritance
System.Object
MultiClassLearner
Inherited Members
System.Object.Equals(System.Object)
System.Object.Equals(System.Object, System.Object)
System.Object.GetHashCode()
System.Object.GetType()
System.Object.MemberwiseClone()
System.Object.ReferenceEquals(System.Object, System.Object)
System.Object.ToString()
Namespace:numl
Assembly:numl.dll
Syntax
public static class MultiClassLearner

Methods

View Source

ChangeClassLabels(Object[], Descriptor, Object)

Returns a Vector of positive and negative labels in 1 - 0 form.

Declaration
public static Vector ChangeClassLabels(object[] examples, Descriptor descriptor, object truthLabel)
Parameters
Type Name Description
System.Object[] examples

Object examples.
Descriptor descriptor

Descriptor.
System.Object truthLabel

The truth label's value (see LabelAttribute).
Returns
Type Description
Vector
View Source

Learn(IGenerator, IEnumerable<Object>, Double, Double, Boolean)

Generate a multi-class classification model using a specialist classifier for each class label.

Declaration
public static ClassificationModel Learn(IGenerator generator, IEnumerable<object> examples, double trainingPercentage, double mixingPercentage = 0.5, bool isMultiClass = true)
Parameters
Type Name Description
IGenerator generator

The generator to use for each individual classifier.
System.Collections.Generic.IEnumerable<System.Object> examples

Training examples of any number of classes
System.Double trainingPercentage

Percentage of training examples to use, i.e. 70% = 0.7
System.Double mixingPercentage

Percentage to mix positive and negative exmaples, i.e. 50% will add an additional 50% of trainingPercentage of negative examples into each classifier when training.
System.Boolean isMultiClass

Determines whether each class is mutually inclusive.

For example: If True, each class takes on a number of classes and does not necessarily belong to one specific class.

The ouput would then be a number of predicted classes for a single prediction. E.g. A song would be True as it may belong to classes: vocals, rock as well as bass.
Returns
Type Description
ClassificationModel