.TH "mlpack::kmeans::RefinedStart" 3 "Tue Sep 9 2014" "Version 1.0.10" "MLPACK" \" -*- nroff -*-
.ad l
.nh
.SH NAME
mlpack::kmeans::RefinedStart \- 
.PP
A refined approach for choosing initial points for k-means clustering\&.  

.SH SYNOPSIS
.br
.PP
.SS "Public Member Functions"

.in +1c
.ti -1c
.RI "\fBRefinedStart\fP (const size_t \fBsamplings\fP=100, const double \fBpercentage\fP=0\&.02)"
.br
.RI "\fICreate the \fBRefinedStart\fP object, optionally specifying parameters for the number of samplings to perform and the percentage of the dataset to use in each sampling\&. \fP"
.ti -1c
.RI "template<typename MatType > void \fBCluster\fP (const MatType &data, const size_t clusters, arma::Col< size_t > &assignments) const "
.br
.RI "\fIPartition the given dataset into the given number of clusters according to the random sampling scheme outlined in Bradley and Fayyad's paper\&. \fP"
.ti -1c
.RI "double \fBPercentage\fP () const "
.br
.RI "\fIGet the percentage of the data used by each subsampling\&. \fP"
.ti -1c
.RI "double & \fBPercentage\fP ()"
.br
.RI "\fIModify the percentage of the data used by each subsampling\&. \fP"
.ti -1c
.RI "size_t \fBSamplings\fP () const "
.br
.RI "\fIGet the number of samplings that will be performed\&. \fP"
.ti -1c
.RI "size_t & \fBSamplings\fP ()"
.br
.RI "\fIModify the number of samplings that will be performed\&. \fP"
.in -1c
.SS "Private Attributes"

.in +1c
.ti -1c
.RI "double \fBpercentage\fP"
.br
.RI "\fIThe percentage of the data to use for each subsampling\&. \fP"
.ti -1c
.RI "size_t \fBsamplings\fP"
.br
.RI "\fIThe number of samplings to perform\&. \fP"
.in -1c
.SH "Detailed Description"
.PP 
A refined approach for choosing initial points for k-means clustering\&. 

This approach runs k-means several times on random subsets of the data, and then clusters those solutions to select refined initial cluster assignments\&. It is an implementation of the following paper:
.PP
{bradley1998refining, title={Refining initial points for k-means clustering}, author={Bradley, Paul S and Fayyad, Usama M}, booktitle={Proceedings of the Fifteenth International Conference on Machine Learning (ICML 1998)}, volume={66}, year={1998} } 
.PP
Definition at line 47 of file refined_start\&.hpp\&.
.SH "Constructor & Destructor Documentation"
.PP 
.SS "mlpack::kmeans::RefinedStart::RefinedStart (const size_tsamplings = \fC100\fP, const doublepercentage = \fC0\&.02\fP)\fC [inline]\fP"

.PP
Create the \fBRefinedStart\fP object, optionally specifying parameters for the number of samplings to perform and the percentage of the dataset to use in each sampling\&. 
.PP
Definition at line 55 of file refined_start\&.hpp\&.
.SH "Member Function Documentation"
.PP 
.SS "template<typename MatType > void mlpack::kmeans::RefinedStart::Cluster (const MatType &data, const size_tclusters, arma::Col< size_t > &assignments) const"

.PP
Partition the given dataset into the given number of clusters according to the random sampling scheme outlined in Bradley and Fayyad's paper\&. 
.PP
\fBTemplate Parameters:\fP
.RS 4
\fIMatType\fP Type of data (arma::mat or arma::sp_mat)\&. 
.RE
.PP
\fBParameters:\fP
.RS 4
\fIdata\fP Dataset to partition\&. 
.br
\fIclusters\fP Number of clusters to split dataset into\&. 
.br
\fIassignments\fP Vector to store cluster assignments into\&. Values will be between 0 and (clusters - 1)\&. 
.RE
.PP

.SS "double mlpack::kmeans::RefinedStart::Percentage () const\fC [inline]\fP"

.PP
Get the percentage of the data used by each subsampling\&. 
.PP
Definition at line 80 of file refined_start\&.hpp\&.
.PP
References percentage\&.
.SS "double& mlpack::kmeans::RefinedStart::Percentage ()\fC [inline]\fP"

.PP
Modify the percentage of the data used by each subsampling\&. 
.PP
Definition at line 82 of file refined_start\&.hpp\&.
.PP
References percentage\&.
.SS "size_t mlpack::kmeans::RefinedStart::Samplings () const\fC [inline]\fP"

.PP
Get the number of samplings that will be performed\&. 
.PP
Definition at line 75 of file refined_start\&.hpp\&.
.PP
References samplings\&.
.SS "size_t& mlpack::kmeans::RefinedStart::Samplings ()\fC [inline]\fP"

.PP
Modify the number of samplings that will be performed\&. 
.PP
Definition at line 77 of file refined_start\&.hpp\&.
.PP
References samplings\&.
.SH "Member Data Documentation"
.PP 
.SS "double mlpack::kmeans::RefinedStart::percentage\fC [private]\fP"

.PP
The percentage of the data to use for each subsampling\&. 
.PP
Definition at line 88 of file refined_start\&.hpp\&.
.PP
Referenced by Percentage()\&.
.SS "size_t mlpack::kmeans::RefinedStart::samplings\fC [private]\fP"

.PP
The number of samplings to perform\&. 
.PP
Definition at line 86 of file refined_start\&.hpp\&.
.PP
Referenced by Samplings()\&.

.SH "Author"
.PP 
Generated automatically by Doxygen for MLPACK from the source code\&.