Data Clustering

Vector Quantization

kmeans(key, points, k[, iter, thresh, max_iters])	Clusters points using k-means algorithm
kmeans_jit(key, points, k[, iter, thresh, …])	Clusters points using k-means algorithm
kmeans_with_seed(key, points, k[, thresh, …])	Runs the k-means algorithm for a specific random initialization
kmeans_with_seed_jit(key, points, k[, …])	Runs the k-means algorithm for a specific random initialization
find_nearest(point, centroids)	Returns the index of the nearest centroid for a specific point
find_nearest_jit(point, centroids)	Returns the index of the nearest centroid for a specific point
find_assignment(points, centroids)	Finds the assignment of each point to a specific centroid
find_assignment_jit(points, centroids)	Finds the assignment of each point to a specific centroid
find_new_centroids(assignment, points, k)	Finds new centroids based on current assignment
find_new_centroids_jit(assignment, points, k)	Finds new centroids based on current assignment

Spectral Clustering

unnormalized(key, W)	Unnormalized spectral clustering
unnormalized_k(key, W, k)	Unnormalized spectral clustering with known number of clusters
unnormalized_k_jit(key, W, k)	Unnormalized spectral clustering with known number of clusters
normalized_random_walk(key, W)	Normalized spectral clustering with random walk

Data types

vq.KMeansState	The state for K-means algorithm
vq.KMeansSolution	The solution for K-means algorithm
spectral.SpectralclusteringSolution	The solution for K-means algorithm