Hi scikit-learn community, I'm interested in reducing a large adjacency matrix (1M x 1M) to 2 dimensions (1M x 2)
This is very similar to force-based graph embedding (drawing): http://en.wikipedia.org/wiki/Force-based_algorithms_(graph_drawing) But instead of using force for optimization, i want to experiment w/ bayesian importance sampling. I have a clear objective function: http://www.smarttypes.org/blog/graph_reduction_linlog_nbody_simulation I want to make this as efficient as possible, i would like to eventually try this in CUDA, but for the immediate future i want to use numpy and the parallelization gained from libblas + liblapack. Curious if anyone can recommend an efficient SIS example? Much, much appreciated, Timmy Wilson Cleveland, OH ------------------------------------------------------------------------------ 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
