The Complete SPRLIB & ANNLIB

pn_adapt_weights

- adapt weights marked ``changed'' according to the accumulated (pseudo-Newton) delta's

SYNOPSIS

int pn_adapt_weights (net, eta, mu, alpha, options)

ARGUMENTS

NET *net A pointer to a NET.
double eta The coefficient eta (the learning rate) of the backpropagation rule.
double alpha The coefficient alpha (the momentum term) of the backpropagation rule.
double mu A small constant used in diverting inflection points in the pseudo-Newton method.
long options See note in pn_adapt_unit.

RETURNS

TRUE if an error was detected, FALSE if no error was detected.

FUNCTION

This function is called only for shared weights networks, since for this type of network first all the gradients have to be known before they can be summed to calculate the actual weight update. Therefore, during the training of shared weights networks, pn_adapt_unit is called for each unit in a certain layer followed by one call to this function, and so on to the first layer.

NOTE

For an explanation of the possible values of options, see the note in pn_adapt_unit.

SEE ALSO

pn_learn

This document was generated using api2html on Thu Mar 5 09:00:00 MET DST 1998