Numbers of points in clusters are not consistent

[Shengchao Ding <di...@gmail.com>]

Hi there,

I’m new to mahout. When I tried to run the example as the instruction, Clustering of synthetic control data, and then extract the results with clusterdumper, I got the results in which  the number shown in the first line, i.e VL-998{n=243 ... }, are not consistent to the actural points displayed in the followed lines (each line is a point), say 242 points. Sometimes the numbers are very different, but sometimes they’re consistent. Most interesting, the sum of the numbers are same, say 600.

I have run this example many times, with mahout 0.5, 0.6 and 0.7, similar errors are shown. Could you figure it out for me?

Thanks,
Shengchao