On Fri, Mar 14, 2014 at 10:38:17AM +0100, Olivier Grisel wrote: > The problem is that by having had a look at the ANN literature in the > past, my gut feeling is that basic LSH is mostly worthless for > practical applications. I would rather not have a method is > scikit-learn that is not performing good enough (in terms of speed up > vs exact brute force queries) to be useful on real data.
Quite clearly. We are not interested in such contribution. Indeed, for my applications, something like random projections would work much better than LSH, as far as I have seen. So what are exactly the usecases and the usage patterns that this proposal is addressing and for which we should see a big speedup? Gaël ------------------------------------------------------------------------------ Learn Graph Databases - Download FREE O'Reilly Book "Graph Databases" is the definitive new guide to graph databases and their applications. Written by three acclaimed leaders in the field, this first edition is now available. Download your free book today! http://p.sf.net/sfu/13534_NeoTech _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
