Clustering package (scipy.cluster)¶
- Hierarchical clustering (scipy.cluster.hierarchy)
- Function Reference
- scipy.cluster.hierarchy.fcluster
- scipy.cluster.hierarchy.fclusterdata
- scipy.cluster.hierarchy.leaders
- scipy.cluster.hierarchy.linkage
- scipy.cluster.hierarchy.single
- scipy.cluster.hierarchy.complete
- scipy.cluster.hierarchy.average
- scipy.cluster.hierarchy.weighted
- scipy.cluster.hierarchy.centroid
- scipy.cluster.hierarchy.median
- scipy.cluster.hierarchy.ward
- scipy.cluster.hierarchy.cophenet
- scipy.cluster.hierarchy.from_mlab_linkage
- scipy.cluster.hierarchy.inconsistent
- scipy.cluster.hierarchy.maxinconsts
- scipy.cluster.hierarchy.maxdists
- scipy.cluster.hierarchy.maxRstat
- scipy.cluster.hierarchy.to_mlab_linkage
- scipy.cluster.hierarchy.dendrogram
- scipy.cluster.hierarchy.ClusterNode
- scipy.cluster.hierarchy.leaves_list
- scipy.cluster.hierarchy.to_tree
- scipy.cluster.hierarchy.is_valid_im
- scipy.cluster.hierarchy.is_valid_linkage
- scipy.cluster.hierarchy.is_isomorphic
- scipy.cluster.hierarchy.is_monotonic
- scipy.cluster.hierarchy.correspond
- scipy.cluster.hierarchy.num_obs_linkage
- References
- Copyright Notice
- Function Reference
- K-means clustering and vector quantization (scipy.cluster.vq)
Vector Quantization / Kmeans¶
Clustering algorithms are useful in information theory, target detection, communications, compression, and other areas. The vq module only supports vector quantization and the k-means algorithms. Development of self-organizing maps (SOM) and other approaches is underway.
Hierarchical Clustering¶
The hierarchy module provides functions for hierarchical and agglomerative clustering. Its features include generating hierarchical clusters from distance matrices, computing distance matrices from observation vectors, calculating statistics on clusters, cutting linkages to generate flat clusters, and visualizing clusters with dendrograms.
