edu.stanford.nlp.cluster
Class HardCluster
java.lang.Object
|
+--edu.stanford.nlp.cluster.SimpleCluster
|
+--edu.stanford.nlp.cluster.HardCluster
- All Implemented Interfaces:
- Cloneable, Cluster
- Direct Known Subclasses:
- HHCluster
- public class HardCluster
- extends SimpleCluster
Hard Cluster. Each datum belongs to only 1 cluster with probability 1.
HardCluster stores term and document probabilities in SparseDoubleArrays. This is more space and time efficient for Hard Clusters, but SoftCluster should be used for clusters which are defined as a smooth probability distribution over terms,documents
Constructor Summary |
HardCluster(int num_terms,
int num_docs)
Allocates memory for arrays and initializes values to 0 |
HardCluster(int num_terms,
int num_docs,
int i)
Allocates memory for arrays, sets index, and initializes values to 0 |
Method Summary |
void |
addDatum(int ind)
Adds a datum to the cluster |
int[] |
dataIndices()
Returns an array of indices of the data that are contained in this cluster |
Methods inherited from class edu.stanford.nlp.cluster.SimpleCluster |
clearData, clone, equals, evaluateIntraSimilarity, evaluateWeightedMean, get_pr_d_z, get_pr_d_z, get_pr_w_z, get_pr_w_z, get_pr_z, getIndex, getIntraSimilarity, getMean, set_pr_d_z, set_pr_d_z, set_pr_w_z, set_pr_w_z, set_pr_z, setIndex, toString, toString, toXMLString, toXMLString |
HardCluster
public HardCluster(int num_terms,
int num_docs)
- Allocates memory for arrays and initializes values to 0
HardCluster
public HardCluster(int num_terms,
int num_docs,
int i)
- Allocates memory for arrays, sets index, and initializes values to 0
dataIndices
public int[] dataIndices()
- Returns an array of indices of the data that are contained in this cluster
addDatum
public void addDatum(int ind)
- Adds a datum to the cluster
Stanford NLP Group