comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » Clustering

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Clustering [message #78202]

Fri, 28 October 2011 05:29

Kai Muehlbauer
Messages: 13
Registered: February 2011

Junior Member

Hi there,

I'm working with multidimensional arrays.

A DOUBLE = Array[2000, 50, 360]

I compute Histograms over the second dimension for all points (2000,360):

tmp_arr = HISTOGRAM( a[K,*,J],BINSIZE=0.1, MAX=15., MIN=0.)

Now I have one Histogram for every point. There are 3 possible peaks in
the histogram output. My intention is to cluster (or group) the
histograms in clusters (or groups) of similar histograms. The maximum
number of groups I would manage in the further processing is 10.

Aim is to finally get an array (2000,360) with contents from 0 to 9.

I stumbled over CLUSTER and CLUST_WTS and tried to get somthing useful
out, but failed so far. For every stripe of 360 histograms I calculated
the weights and CLUSTER, but this did not seem the right approach.

I appreciate any hints and suggestions.

Cheers,

Kai

Report message to a moderator

[Message index]

Clustering

By: Kai Muehlbauer on Fri, 28 October 2011 05:29

Previous Topic:	questions regarding using cgWindow, cgLoadCT, and something else
Next Topic:	Re: Alert IDL

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 15:32:47 PDT 2025

Total time taken to generate the page: 0.00582 seconds