Fn_kmeans

Functions
template<typename T1 >
void	ecocpak::kmeans (Col< u32 > &indices_out, Col< u32 > &ranks_out, Mat< typename T1::elem_type > &centroids_out, const Base< typename T1::elem_type, T1 > &X, const u32 k, const typename T1::pod_type lr=0.01)
mat	ecocpak::subClasses (ucolvec &inds, ucolvec &ranks, const mat &samples)

---

Function Documentation

template<typename T1 >

void ecocpak::kmeans	(	Col< u32 > &	indices_out,
		Col< u32 > &	ranks_out,
		Mat< typename T1::elem_type > &	centroids_out,
		const Base< typename T1::elem_type, T1 > &	X,
		const u32	k,
		const typename T1::pod_type	lr = 0.01
	)		[inline]

• K-means clustering.

• Input Arguments:
  • X : Base object which includes the samples' matrix
  • k : k denotes the number of desired clusters.
  • lr : Starting learning rate for online phase version of K-means (default 1%).

• Output Arguments:
  • indices_out : Cluster indices for each point.
  • ranks_out : Vector which holds the number of samples that correspond to each centroid.
  • centroids_out : Cluster centroid locations.

• Return Argument:
  • Void.

mat ecocpak::subClasses	(	ucolvec &	inds,
		ucolvec &	ranks,
		const mat &	samples
	)

• Function subClasses splits a data set in two subclasses using K-means clustering.

• Input Arguments:
  • inds : Sample cluster indices.
  • ranks : A vector which its position denotes the number of samples in each cluster.
  • samples : Sample vectors' matrix.

• Output Arguments:
  • Void.

• Return Argument:
  • A matrix which represents the centers of clusters.