MPIKmeanshttp://mloss.orgUpdates and additions to MPIKmeansenThu, 24 Jun 2010 14:42:57 -0000MPIKmeans 1.5<html><p>A K-means clustering implementation for command-line, Python, Matlab and C. This algorithm yields the very same solution as standard Kmeans, even after each iteration. However it uses some triangle inequalities and stores some distances. Therefore it is MUCH faster than standard Kmeans but uses more memory. See the corresponding paper for more details if you are interested. </p></html>Peter GehlerFri, 16 Jan 2009 15:48:47 -0000<b>Comment by Dmitrey Kroshko on 2008-09-02 20:20</b><p>I try to run makefile (from linux KUBUNTU) but it seems like it requires MATLAB installed. Is it true?</p> Dmitrey KroshkoTue, 02 Sep 2008 20:20:15 -0000<b>Comment by Peter Gehler on 2008-09-03 09:12</b><p>You can als run "make shared" and use the python version or bind your own C code.</p> Peter GehlerWed, 03 Sep 2008 09:12:40 -0000<b>Comment by Jerry on 2008-10-16 14:18</b><p>"help mpi<em>kmeans</em>mex" does not work! I think there is no mpi<em>kmeans</em>mex.m file in the Archive:)</p> JerryThu, 16 Oct 2008 14:18:22 -0000<b>Comment by wang haoxue on 2008-11-25 02:43</b><p>this means is very good!</p> wang haoxueTue, 25 Nov 2008 02:43:11 -0000<b>Comment by zzinwhu on 2009-02-15 19:49</b><p>How can I download it?</p> zzinwhuSun, 15 Feb 2009 19:49:23 -0000<b>Comment by Bill on 2009-03-21 04:04</b><p>Hello! I need to use this function for my research experiment in matlab. But the archive was missing file "mpi<em>kmeans</em>mex", so I couldn't run "help mpi<em>kmeans</em>mex". please help me! thx!</p> BillSat, 21 Mar 2009 04:04:41 -0000<b>Comment by Peter Gehler on 2009-04-07 16:04</b><p>There was a typo in the README. To get help in matlab simply type "help mpi_kmeans"</p> Peter GehlerTue, 07 Apr 2009 16:04:58 -0000<b>Comment by Janosch Peters on 2010-06-24 14:42</b><p>mpi kmeans is causing a segfault when the number of clusters is too close to the number of features. See the gdb output below:</p> <pre><code>Program received signal SIGSEGV, Segmentation fault. 0x00007fffef49e465 in add_point_to_cluster (cluster_ind=71796994, CX=0x8f2eb70, px=0xb90beb0, nr_points=0x6c235a0, dim=128) at /home/zaphod/Code/Recognosco/src/mpi_kmeans-1.5/mpi_kmeans.cxx:100 100 if (nr_points[cluster_ind]==0) (gdb) backtrace #0 0x00007fffef49e465 in add_point_to_cluster (cluster_ind=71796994, CX=0x8f2eb70, px=0xb90beb0, nr_points=0x6c235a0, dim=128) at /home/zaphod/Code/Recognosco/src/mpi_kmeans-1.5/mpi_kmeans.cxx:100 #1 0x00007fffef49e68b in remove_identical_clusters (CX=0x8f2eb70, cluster_distance=0xbad9ac0, X=0xa8706b0, cluster_count=0x6c235a0, c=0x6b1ea80, dim=128, nclus=2000, npts=18853) at /home/zaphod/Code/Recognosco/src/mpi_kmeans-1.5/mpi_kmeans.cxx:138 #2 0x00007fffef49f12f in kmeans_run (CX=0x8f2eb70, X=0xa8706b0, c=0x6b1ea80, dim=128, npts=18853, nclus=2000, maxiter=0) at /home/zaphod/Code/Recognosco/src/mpi_kmeans-1.5/mpi_kmeans.cxx:373 #3 0x00007fffef49fc78 in kmeans (CX=0x8f2eb70, X=0xa8706b0, assignment=0x6b1ea80, dim=128, npts=18853, nclus=2000, maxiter=0, restarts=20) at /home/zaphod/Code/Recognosco/src/mpi_kmeans-1.5/mpi_kmeans.cxx:584 </code></pre> Janosch PetersThu, 24 Jun 2010 14:42:57 -0000