úθ)      !"#$%&'(portable experimentalrobert.steuck@gmail.com) the neuron input weights activation function ,first derivation of the activation function 3information generated during a simple forward pass  output sum of weighted inputs inputs 1/ (1+e^(-x)) first derivation activation function is ) number of inputs !seed for random weigth generator number of neurons per layer )generate forward pass info for a network 'generate forward pass info for a layer (generate forward pass info for a neuron *calculate the weigtet input of the neuron )updates the weigts for an entire network learning rate alpha desired output value updates the weigts for a layer ! updates the weigts for a neuron "learning rate alpha inpit and desired output #<calculates the output of a network for a given input vector $:calculates the output of a layer for a given input vector %)quadratic error for a single vector pair &'quadratic error for for multiple pairs ',produces an indefinite sequence of networks learning rate alpha *list of pairs of input and desired output (8trains a network with a set of vector pairs until a the & is smaller than epsilon learning rate alpha the maximum error epsilon *list of pairs of input and desired output -  !"#$%&'(*+,-)   !"#$%&'()   !"#$%&'(.      !"#$%&'()*+,-./01 bpann-0.1.1AI.BPANN NeuronCreator PackedNeuronNeuronwsfunfun'ForwardPassInfoFPInfoonetxsNetworkANetworkALayersigmoidsigmoid' sigmoidNeuron outputNeuron biasNeuron createLayer sigmoidLayer outputLayercreateRandomNetworkicNcToPackedNeurons unpackNetwork packNetwork passForward passForward' passForward''calcNet weightUpdate weightUpdate'weightUpdate''backprop calculate calculate' quadErrorNetglobalQuadError trainAlottrainbaseGHC.Baseid testBoolAnd testBoolOr testBoolXor testBoolNot