I am working on a regression problem. Currently I'm using pybrain with a classic neural net approach.
I iterated over some number (100) of trials. For each trial, I generate some number (20000) training vectors. The training is "online", in the sense that I feed 20000 vectors, evaluate the accuracy of the NN, then another 20000 training vectors, ... With each trial, the learning is _cumulative_, I'm not starting over with a random NN, but with each batch of vectors improving the learning (hopefully). I wonder if scikit-learn could be used in a similar manner? ------------------------------------------------------------------------------ Open source business process management suite built on Java and Eclipse Turn processes into business applications with Bonita BPM Community Edition Quickly connect people, data, and systems into organized workflows Winner of BOSSIE, CODIE, OW2 and Gartner awards http://p.sf.net/sfu/Bonitasoft _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
