cIsdbscan: cIsdbscan

In InfOmics/ISDBSCAN-R: Clustering Large Single-Cell Datasets

Description

cIsdbscan for both matrix and HDF5Matrix

Usage

cIsdbscan(data, k, batch_size, stratif)

Arguments

data	numeric matrix or integer matrix or HDF5Matrix
k	the number of clusters
batch_size	the size of the mini batches

Details

This function performs k-means clustering using mini batches.

kmeans++: kmeans++ initialization. Reference : http://theory.stanford.edu/~sergei/papers/kMeansPP-soda.pdf AND http://stackoverflow.com/questions/5466323/how-exactly-does-k-means-work

random: random selection of data rows as initial centroids

Value

a list with the following attributes: centroids, WCSS_per_cluster, best_initialization, iters_per_initialization

References

https://github.com/mlampros/ClusterR

InfOmics/ISDBSCAN-R documentation built on May 30, 2019, 2:04 a.m.