Class to represent a partitioning of a data set. More...
#include <cstddef>
#include <vector>
#include <set>
#include <ostream>
#include <stdint.h>
Go to the source code of this file.
Classes | |
struct | partition |
This represents a partitioning of a data set. More... | |
struct | partition::member_writer |
writable structure returned by members() function. More... | |
Namespaces | |
namespace | cluster |
Namespace for everything in the cluster library. | |
Typedefs | |
typedef size_t | medoid_id |
More descriptive type for medoid index. | |
typedef size_t | object_id |
More descriptive type for object index. | |
typedef std::vector< std::set < object_id > > | cluster_list |
Explicit representation of a clustering. | |
Functions | |
std::ostream & | operator<< (std::ostream &out, const partition::member_writer &mw) |
std::ostream & | operator<< (std::ostream &out, const cluster_list &list) |
Prints out nicely formatted clustering. | |
std::ostream & | operator<< (std::ostream &out, const partition &km) |
For convenience. | |
double | mirkin_distance (const cluster_list &c1, const cluster_list &c2) |
Mirkin distance bt/w two clusterings. | |
double | mirkin_distance (const partition &c1, const partition &c2) |
Convenience overload for comparing partition objects directly. | |
void | expand (cluster_list &list, size_t level=1) |
Expand a cluster_list by l levels. | |
template<typename D > | |
double | total_dissimilarity (const partition &p, D dist) |
Compute the total dissimilarity between all objects and their medoids. | |
template<typename D > | |
double | total_dissimilarity (const partition &p, D dist, medoid_id m) |
Compute the total dissimilarity between all objects in a particular cluster and its medoid. | |
template<typename D > | |
double | total_squared_dissimilarity (const partition &p, D dist) |
Compute the total squared dissimilarity between all objects and their medoids. | |
template<typename D > | |
double | total_squared_dissimilarity (const partition &p, D dist, medoid_id m) |
Compute the total squared dissimilarity between all objects in a particular cluster and its medoid. |
Class to represent a partitioning of a data set.
Definition in file partition.h.