Class FeatureLearning

This class is designed to learn bases (cluster centroids) through feature learning process.

Inheritance
Object
FeatureLearning
Inherited Members
Namespace: AudioAnalysisTools.DSP
Assembly: AudioAnalysisTools.dll
Syntax
public static class FeatureLearning

Methods

| Improve this Doc View Source

MaxPooling(Double[,], Int32)

This method downsamples the input matrix (x,y) by a factor of n on the temporal scale (x) using max pooling.

Declaration
public static double[, ] MaxPooling(double[, ] matrix, int factor)
Parameters
Type Name Description
Double[,] matrix
Int32 factor
Returns
Type Description
Double[,]
| Improve this Doc View Source

SemisupervisedFeatureLearning(FeatureLearningSettings, String, String[,])

This method is called semi-supervised feature learning because one of the clusters is formed using the positive frames manually selected from 1-min recordings. The input to this methods is a group of files that contains the call of interest, a 2D-array that contains file name, the second number and the corresponding frame numbers in each file. At the moment, this method only handles single-frames as patches (PatchHeight = 1).

Declaration
public static List<KmeansClustering.Output> SemisupervisedFeatureLearning(FeatureLearningSettings config, string inputPath, string[, ] frameInfo)
Parameters
Type Name Description
FeatureLearningSettings config
String inputPath
String[,] frameInfo
Returns
Type Description
List<KmeansClustering.Output>
| Improve this Doc View Source

UnsupervisedFeatureLearning(FeatureLearningSettings, String)

Apply feature learning process on a set of patch sampling set in an unsupervised manner Output clusters.

Declaration
public static List<KmeansClustering.Output> UnsupervisedFeatureLearning(FeatureLearningSettings config, string inputPath)
Parameters
Type Name Description
FeatureLearningSettings config
String inputPath
Returns
Type Description
List<KmeansClustering.Output>