How long would it take an SVM to train on a 12 input, 1000 or 2000 row dataset? How about a Neural Network? Would the neural network train faster or slower than SVM? Assume that it would be running on a modern quad-core i7. I don't have data yet, just an idea...wondering if it's feasible. By the way, are there any plans for RapidMiner to support OpenCL or GPU computing?