Sets training method and common parameters

model.setTrainMethod(method)
model.setTrainMethod(method, 'OptionName', optionValue, ...)

Input

• method Available training methods:
  • Backprop The back-propagation algorithm.
  • RProp (default) The RPROP algorithm. See [RPROP93] for details.
  • Anneal The simulated annealing algorithm. See [Kirkpatrick83] for details.

Options

• Param1 sets RpropDW0 property for 'RProp' and sets BackpropWeightScale property for 'Backprop' and sets AnnealInitialT for Anneal. default 0
• Param2 sets RpropDWMin property for 'RProp' and sets BackpropMomentumScale property for 'Backprop' and sets AnnealFinalT for Anneal. default 0

References

[RPROP93]:

Martin Riedmiller and Heinrich Braun. "A direct adaptive method for faster backpropagation learning: The rprop algorithm". In Neural Networks, 1993., IEEE International Conference on, pages 586-591. IEEE, 1993.

[Kirkpatrick83]:

S. Kirkpatrick, C. D. Jr Gelatt, and M. P. Vecchi. "Optimization by simulated annealing". Science, 220(4598):671-680, 1983.