quickProp [<num-updates>] [-report <report-interval>]
This is a shortcut for training the network using quick-prop. The arguments are similar to those for train. The momentum term is taken from the network's momentum parameter.
This algorithm assumes that the error surface in weight space is parabolic and tries to estimate the minimum point based on the curvature. It then jumps towards the estimated minumum. The learningRate gives the proportion of the distance towards the minimum it will jump. This algorithm tends to be very unstable.
To train for 1000 epochs, printing reports every 100:
lens> quickProp 1000 -r 100