On 2012-03-07, at 8:45 AM, Olivier Grisel wrote: > I think was DWF calls sparse coding is the LASSO implemented with > coordinate descent (sparse coding with a fixed dictionary).
Indeed, the encoding step rather than the dictionary learning step. When benchmarking unsupervised feature learning methods it has become common to separately consider the learning step from the encoding step, as it seems the latter is a good deal more important, at least for classification accuracy (see http://www.stanford.edu/~acoates/papers/coatesng_icml_2011.pdf ). Right now doing that step on a large scale with scikit-learn is somewhat infeasible. I'll try running it through line_profiler today and see what I come up with. David ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
