资 源 简 介
Usage: hadoop jar random-seed-generator-0.1.0.jar org.mathbiol.mahout.RandomSeedGenerator
This Hadoop tool is a spin-off of the RandomSeedGenerator in Mahout"s implementation of k-means clustering that creates k random initial seeds from vectorized data.
refers to the vectorized data, and to a directory where the random clusters should be stored. The number of clusters can be specified with . Only if some other distance measure than SquaredEuclideanDistanceMeasure is used, the full path has to be specified in .
