On Fri, Oct 05, 2012 at 09:56:00AM +0200, Gael Varoquaux wrote: > The reason that your Lars implementation does not suffer from this > problem, and probably the reason that the R implementation does not > either, is most likely that our Lars implementation iteratively refines > the computation of the residuals and of the Cholesky gram matrix.
Actually, I was wrong with regards to the computation of the residuals: if you look at 'least_angle.py', line 335, we are computing the residuals using the full computation, and there is a 'TODO' stressing that they could be updated. On the other hand, we are definitely refining the computation of the Cholesky while you are using a 'linalg.solve'. Our implementation is clearly faster, and more prone to numerical errors. Gaël ------------------------------------------------------------------------------ Don't let slow site performance ruin your business. Deploy New Relic APM Deploy New Relic app performance management and know exactly what is happening inside your Ruby, Python, PHP, Java, and .NET app Try New Relic at no cost today and get our sweet Data Nerd shirt too! http://p.sf.net/sfu/newrelic-dev2dev _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
