> I transform my data using different transformation T1 and T2 and then feed > it into the LInearSVC. What I found is that the classifier is trained > significantly faster with the transformed data using T2. > > Since both transformed data has the same number of instances, we are looking > at factors other than number of instances that affect the training time. So > my question is basically what are the properties of a dataset that make the > LinearSVC training time shorter?
LinearSVC converges faster with high regularization parameter (small C) and the speed depends on the conditioning of the problem. If you had orthogonal features it would be super fast. The more correlated they are the slower it is. HTH Alex ------------------------------------------------------------------------------ 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 Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
