spark-scala-KMeans

Here is a basic KMeans algorithm in spark/scala. It uses arrays - it makes no use of the MLLIB. If you are in a hurry and want to develop on something basic, then this can be a good statrting point. The input txt file is just a tab delimited txt file. Each row in this text file represents a point. Set numClusters (number of clusters you'd like to have) and numIterations before using it. At the end, this code just prints out the centers and the center to which each point is mapped to. Please let me know if you have amy questions ([email protected]).