I've seen some simple examples of neural networks on Github but nothing seems to be complete and/or reliable. If I had infinite time, I would like to experiment with ForwardDiff.jl to do automatic differentiation of expressions to generate backprop expressions for arbitrary MLP architectures (something similar to what Theano does, even though Theano does much more than that).
-- João Felipe Santos On Sat, Jun 7, 2014 at 12:11 PM, John Myles White <[email protected]> wrote: > I don’t think there’s any reliable neural network Julia code published so > far. > > — John > > On Jun 7, 2014, at 1:18 AM, [email protected] wrote: > > > > > I am just wondering if there is code for multi layer perceptron anywhere > in julia with parallel processing and iRprop+ for gradient decent? > >
