Wrap your input vectors in a NamedVector and they will pass through the clustering intact. Use the name to re-associate the other fields with your clustered vector.
-----Original Message----- From: dbg [mailto:[email protected]] Sent: Thursday, July 14, 2011 1:50 PM To: [email protected] Subject: Fields needed after clustering but not used within Mahout Caveat: New Mahout user + New Clustering Developer I am running kmeans/hadoop on a data set with fields that are needed after clustering but are not part of the vectors used to calculate the vectors. Creating a map to get this data back after clustering won't work because the fields sent to Mahout aren't necessarly unique: 1,2,3,4,5,6,BOB 1,2,3,4,5,6,PHIL 2,3,3,4,5,6,Cindy Is there a mechanism around this? -- View this message in context: http://lucene.472066.n3.nabble.com/Fields-needed-after-clustering-but-not-used-within-Mahout-tp3170297p3170297.html Sent from the Mahout Developer List mailing list archive at Nabble.com.
