This command trains a backpropagation network using the RPROP training algorithm. The method is based on adaptive learning rate parameters on each neuron's weight. By default one neuron for each input is used.
Example (ex5.11): Train a three-layered (input + hidden + output layer) network with a sine function using a sigmoid activation function in neurons. After training is done save the network output and error.
NDA> load sink.dat NDA> load sint.dat NDA> select sinx -f sink.ox NDA> select siny -f sink.oy NDA> select sinox -f sink.ox NDA> select sintx -f sint.tx NDA> select sinty -f sint.ty NDA> rprop -di sinx -do siny -net 3 1 3 1 -types s s s -em 40 -bs 0.05 -mup 1.1 -mdm 0.8 -nout wei -ti sintx -to sinty -ef virhe NDA> fbp -di sinox -do out -win wei NDA> select test -f virhe.TrainError virhe.TestError NDA> select output -f sink.ox out.0 NDA> save output NDA> save test
