https://sourceforge.net/p/python-cluster/news/2013-03-11T13:05:08ZRecent posts to news2013-03-11T13:05:08Z2013-03-11T13:05:08ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.neta8f64915bef83b4d5332686ef493d1e082c8bdbd

<div class="markdown_content"><p>Fixed bug caused by numpy arrays and KMeansClustering.</p> <p>The KCuster constructor now accepts an optional function to test for item equality. If the clusters contain numpy arrays, you can pass "numpy.array_equals".</p></div>

2013-03-11T13:03:34Z2013-03-11T13:03:34ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.net92b14ed79bb1107daa8f4065ab616fa5e83d308d

<div class="markdown_content"><p>The project is now available in github here:<br /> <a href="http://www.github.com/exhuma/python-cluster" rel="nofollow">http://www.github.com/exhuma/python-cluster</a></p> <p>The news-feed and project page will remain on sourceforge for now.</p></div>

2007-10-14T10:18:56Z2007-10-14T10:18:56ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.netd13135c3a0742e876cafd0be2fe0b305c03364ab

<div class="markdown_content"><p>Finally migrated to SVN (https://sourceforge.net/svn/?group_id=170665).</p> <p>I did not bother about migrating the history as well. Not really necessary for such a small project. I will leave CVS access enabled for a while but will take it down eventually. So if you pull from CVS, make sure that you switch to the SVN repo as soon as possible.</p></div>

2007-05-24T07:15:25Z2007-05-24T07:15:25ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.netb86044e171eef242feed39125e20b30c826d6fa9

<div class="markdown_content"><p>A helpful hand discovered a broken link in the source.</p> <p>Broken link: <a href="http://mail.python.org/pipermail/python-list/2004-December/253517.html" rel="nofollow">http://mail.python.org/pipermail/python-list/2004-December/253517.html</a></p> <p>Alternative link: <a href="http://mail.python.org/pipermail/python-list/2004-December/294990.html" rel="nofollow">http://mail.python.org/pipermail/python-list/2004-December/294990.html</a></p> <p>It's not the same link, but contains similar information. I hope this link will stay.</p></div>

2006-08-19T11:33:00Z2006-08-19T11:33:00ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.net720faa247db2d141cd3b9f3b6ce760593294bdb4

<div class="markdown_content"><p>1.1.1b1<br /> - Applied patch [1535137] (thanks ajaksu)<br /> --&gt; Topology output supported<br /> --&gt; data and raw_data are now properties.</p></div>

2006-07-12T14:18:53Z2006-07-12T14:18:53ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.net77141b3e24575de3d02806a353899bdb4fa21940

<div class="markdown_content"><p>K-Means clustering is now implemented</p></div>

2006-07-06T11:42:31Z2006-07-06T11:42:31ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.net8a6a2935ca3312d9b894e9eabcb60eab206f29e8

<div class="markdown_content"><p>This is a bug-fix release!<br /> It fixes bug #1516204 that caused the clustering algorithm to raise an exception if an empty list or a list of only one item was supplied as argument.</p> <p>In addition I added some unit-tests. This keeps development and bug-tracking a lot easier.</p></div>

2006-07-05T06:49:45Z2006-07-05T06:49:45ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.net23de22c67e4fdb9c122e19615fba460be1e41a8a

<div class="markdown_content"><p>Eeeks. The last header in the news read &quot;new release for python-ngram&quot;. Although that is somewhat true, it really should have read &quot;python-cluster&quot;.</p> <p>Sorry.</p></div>

2006-07-05T06:47:54Z2006-07-05T06:47:54ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.netb69321150717541e3270d3d8460e4e0f9cb197fb

<div class="markdown_content"><p>Finally I got around to build the dist-files and upload them. Enjoy.</p> <p>You might notice that there are only the source distribution and the windows binary distribution. I decided to drop the rest because they are in my opinion unneeded. They all behave exactly the same as the source distribution anyway. Yes, I could have created the other files as well for convenience, but how hard is it really to type &quot;python setup.py install&quot;? ;)</p> <p>And yes, as the windows-people like the click-and-play behaviour, I kept the .exe file.</p> <p>The other reason behind this, is that it's awkward to release multiple files on sourceforge. For *every* ... *single* ... *file* you have to select the settings and refresh the page. After the first few releases, it really becomes tedious and repetitive.</p> <p>- End.</p></div>

2006-07-03T16:51:07Z2006-07-03T16:51:07ZMichel Alberthttps://sourceforge.net/u/exhuma/https://sourceforge.netd5770bf6a9b6349bdb56db36cbad2c737ce98514

<div class="markdown_content"><p>I finished a new version today. Now the hierarchical clustering works twice(!) as fast. This is achieved because the distance-matrix which is generated internally is symmetric. So I only need to calculate one half of the possible combinations.</p> <p>More improvement is possible. But I gave up on that today. Too complex.</p> <p>I also started to work out the details for a K-Means algorithm on the Airport this weekend. On paper it looks sensible and I beleive it should work as I wrote it down. Now I only need to beam my scribblings onto the harddisk. And hope it all works ;)</p></div>