2013/9/8 Gael Varoquaux <gael.varoqu...@normalesup.org>: > On Sun, Sep 08, 2013 at 05:14:35PM +0200, Alexandre Gramfort wrote: >> I would be in favor of a HierarchicalClustering object that supports >> various linkage >> criteria. > >> something like: > >> hc = HierarchicalClustering(linkage='single') > >> linkage='ward' would be another option. > > Yes, indeed. This is what I am working on in my branch. I was planning to > try to merge this in with Ward, complete and average linkage, as single > linkage is best implemented with a different algorithm (indeed, an MST).
Note: in Gael's branch this is implemented as `AgglomerativeClustering` to be more specific as `HierarchicalClustering` could also be based on divisive strategies would might have a different family of hyper parameters. > The branch still requires finishing work to get things like arbitrary > metrics supported, as well as documentation, but the core of the work in > done. > > https://github.com/scikit-learn/scikit-learn/pull/2199 Robert, could you have a look at this branch to better have an idea of the overlap between this branch and your work and how much would it benefit by reusing common datastructures / classes / utility functions? -- Olivier http://twitter.com/ogrisel - http://github.com/ogrisel ------------------------------------------------------------------------------ Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! Discover the easy way to master current and previous Microsoft technologies and advance your career. Get an incredible 1,500+ hours of step-by-step tutorial videos with LearnDevNow. Subscribe today and save! http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
