|
kmeans
k-means clustering in C++
|
Additional statistics from the k-means algorithm. More...
#include <Details.hpp>
Public Attributes | |
| std::vector< Index_ > | sizes |
| int | iterations = 0 |
| int | status = 0 |
Additional statistics from the k-means algorithm.
| Index_ | Integer type of the observation index. |
| int kmeans::Details< Index_ >::iterations = 0 |
The number of iterations that were performed. This can be interpreted as the number of iterations to convergence if Details::status == 0.
| std::vector<Index_> kmeans::Details< Index_ >::sizes |
The number of observations in each cluster. This has length equal to num_centers in Refine::run(), where sizes[j] contains the number of observations in cluster j.
Some clusters may be empty, e.g., when there are more requested centers than clusters - see remove_unused_centers() to optionally remove such clusters.
| int kmeans::Details< Index_ >::status = 0 |
The status of the algorithm on completion. A value of 0 indicates that the algorithm completed successfully. The interpretation of a non-zero value depends on the algorithm.