MLPACK
1.0.7
|
This class implements K-Means clustering. More...
Public Member Functions | |
KMeans (const size_t maxIterations=1000, const double overclusteringFactor=1.0, const MetricType metric=MetricType(), const InitialPartitionPolicy partitioner=InitialPartitionPolicy(), const EmptyClusterPolicy emptyClusterAction=EmptyClusterPolicy()) | |
Create a K-Means object and (optionally) set the parameters which K-Means will be run with. More... | |
template<typename MatType > | |
void | Cluster (const MatType &data, const size_t clusters, arma::Col< size_t > &assignments, const bool initialGuess=false) const |
Perform k-means clustering on the data, returning a list of cluster assignments. More... | |
template<typename MatType > | |
void | Cluster (const MatType &data, const size_t clusters, arma::Col< size_t > &assignments, MatType ¢roids, const bool initialAssignmentGuess=false, const bool initialCentroidGuess=false) const |
Perform k-means clustering on the data, returning a list of cluster assignments and also the centroids of each cluster. More... | |
const EmptyClusterPolicy & | EmptyClusterAction () const |
Get the empty cluster policy. More... | |
EmptyClusterPolicy & | EmptyClusterAction () |
Modify the empty cluster policy. More... | |
template<typename MatType > | |
void | FastCluster (MatType &data, const size_t clusters, arma::Col< size_t > &assignments) const |
An implementation of k-means using the Pelleg-Moore algorithm; this is known to not work – do not use it! (Fixing it is TODO, of course; see #251.) More... | |
size_t | MaxIterations () const |
Get the maximum number of iterations. More... | |
size_t & | MaxIterations () |
Set the maximum number of iterations. More... | |
const MetricType & | Metric () const |
Get the distance metric. More... | |
MetricType & | Metric () |
Modify the distance metric. More... | |
double | OverclusteringFactor () const |
Return the overclustering factor. More... | |
double & | OverclusteringFactor () |
Set the overclustering factor. Must be greater than 1. More... | |
const InitialPartitionPolicy & | Partitioner () const |
Get the initial partitioning policy. More... | |
InitialPartitionPolicy & | Partitioner () |
Modify the initial partitioning policy. More... | |
Private Attributes | |
EmptyClusterPolicy | emptyClusterAction |
Instantiated empty cluster policy. More... | |
size_t | maxIterations |
Maximum number of iterations before giving up. More... | |
MetricType | metric |
Instantiated distance metric. More... | |
double | overclusteringFactor |
Factor controlling how many clusters are actually found. More... | |
InitialPartitionPolicy | partitioner |
Instantiated initial partitioning policy. More... | |
This class implements K-Means clustering.
This implementation supports overclustering, which means that more clusters than are requested will be found; then, those clusters will be merged together to produce the desired number of clusters.
Two template parameters can (optionally) be supplied: the policy for how to find the initial partition of the data, and the actions to be taken when an empty cluster is encountered, as well as the distance metric to be used.
A simple example of how to run K-Means clustering is shown below.
MetricType | The distance metric to use for this KMeans; see metric::LMetric for an example. |
InitialPartitionPolicy | Initial partitioning policy; must implement a default constructor and 'void Cluster(const arma::mat&, const size_t, arma::Col<size_t>&)'. |
EmptyClusterPolicy | Policy for what to do on an empty cluster; must implement a default constructor and 'void EmptyCluster(const arma::mat&, arma::Col<size_t&)'. |
Definition at line 75 of file kmeans.hpp.
mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::KMeans | ( | const size_t | maxIterations = 1000 , |
const double | overclusteringFactor = 1.0 , |
||
const MetricType | metric = MetricType() , |
||
const InitialPartitionPolicy | partitioner = InitialPartitionPolicy() , |
||
const EmptyClusterPolicy | emptyClusterAction = EmptyClusterPolicy() |
||
) |
Create a K-Means object and (optionally) set the parameters which K-Means will be run with.
This implementation allows a few strategies to improve the performance of K-Means, including "overclustering" and disallowing empty clusters.
The overclustering factor controls how many clusters are actually found; for instance, with an overclustering factor of 4, if K-Means is run to find 3 clusters, it will actually find 12, then merge the nearest clusters until only 3 are left.
maxIterations | Maximum number of iterations allowed before giving up (0 is valid, but the algorithm may never terminate). |
overclusteringFactor | Factor controlling how many extra clusters are found and then merged to get the desired number of clusters. |
metric | Optional MetricType object; for when the metric has state it needs to store. |
partitioner | Optional InitialPartitionPolicy object; for when a specially initialized partitioning policy is required. |
emptyClusterAction | Optional EmptyClusterPolicy object; for when a specially initialized empty cluster policy is required. |
void mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::Cluster | ( | const MatType & | data, |
const size_t | clusters, | ||
arma::Col< size_t > & | assignments, | ||
const bool | initialGuess = false |
||
) | const |
Perform k-means clustering on the data, returning a list of cluster assignments.
Optionally, the vector of assignments can be set to an initial guess of the cluster assignments; to do this, set initialGuess to true.
MatType | Type of matrix (arma::mat or arma::sp_mat). |
data | Dataset to cluster. |
clusters | Number of clusters to compute. |
assignments | Vector to store cluster assignments in. |
initialGuess | If true, then it is assumed that assignments has a list of initial cluster assignments. |
void mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::Cluster | ( | const MatType & | data, |
const size_t | clusters, | ||
arma::Col< size_t > & | assignments, | ||
MatType & | centroids, | ||
const bool | initialAssignmentGuess = false , |
||
const bool | initialCentroidGuess = false |
||
) | const |
Perform k-means clustering on the data, returning a list of cluster assignments and also the centroids of each cluster.
Optionally, the vector of assignments can be set to an initial guess of the cluster assignments; to do this, set initialAssignmentGuess to true. Another way to set initial cluster guesses is to fill the centroids matrix with the centroid guesses, and then set initialCentroidGuess to true. initialAssignmentGuess supersedes initialCentroidGuess, so if both are set to true, the assignments vector is used.
Note that if the overclustering factor is greater than 1, the centroids matrix will be resized in the method. Regardless of the overclustering factor, the centroid guess matrix (if initialCentroidGuess is set to true) should have the same number of rows as the data matrix, and number of columns equal to 'clusters'.
MatType | Type of matrix (arma::mat or arma::sp_mat). |
data | Dataset to cluster. |
clusters | Number of clusters to compute. |
assignments | Vector to store cluster assignments in. |
centroids | Matrix in which centroids are stored. |
initialAssignmentGuess | If true, then it is assumed that assignments has a list of initial cluster assignments. |
initialCentroidGuess | If true, then it is assumed that centroids contains the initial centroids of each cluster. |
|
inline |
Get the empty cluster policy.
Definition at line 191 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::emptyClusterAction.
|
inline |
Modify the empty cluster policy.
Definition at line 194 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::emptyClusterAction.
void mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::FastCluster | ( | MatType & | data, |
const size_t | clusters, | ||
arma::Col< size_t > & | assignments | ||
) | const |
An implementation of k-means using the Pelleg-Moore algorithm; this is known to not work – do not use it! (Fixing it is TODO, of course; see #251.)
|
inline |
Get the maximum number of iterations.
Definition at line 176 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::maxIterations.
|
inline |
Set the maximum number of iterations.
Definition at line 178 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::maxIterations.
|
inline |
Get the distance metric.
Definition at line 181 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::metric.
|
inline |
Modify the distance metric.
Definition at line 183 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::metric.
|
inline |
Return the overclustering factor.
Definition at line 171 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::overclusteringFactor.
|
inline |
Set the overclustering factor. Must be greater than 1.
Definition at line 173 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::overclusteringFactor.
|
inline |
Get the initial partitioning policy.
Definition at line 186 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::partitioner.
|
inline |
Modify the initial partitioning policy.
Definition at line 188 of file kmeans.hpp.
References mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::partitioner.
|
private |
Instantiated empty cluster policy.
Definition at line 206 of file kmeans.hpp.
Referenced by mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::EmptyClusterAction().
|
private |
Maximum number of iterations before giving up.
Definition at line 200 of file kmeans.hpp.
Referenced by mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::MaxIterations().
|
private |
Instantiated distance metric.
Definition at line 202 of file kmeans.hpp.
Referenced by mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::Metric().
|
private |
Factor controlling how many clusters are actually found.
Definition at line 198 of file kmeans.hpp.
Referenced by mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::OverclusteringFactor().
|
private |
Instantiated initial partitioning policy.
Definition at line 204 of file kmeans.hpp.
Referenced by mlpack::kmeans::KMeans< MetricType, InitialPartitionPolicy, EmptyClusterPolicy >::Partitioner().