wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual gbayes wolffd@0: wolffd@0: wolffd@0: wolffd@0:

gbayes wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Evaluate gradient of Bayesian error function for network. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0:

wolffd@0: g = gbayes(net, gdata)
wolffd@0: [g, gdata, gprior] = gbayes(net, gdata)
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: g = gbayes(net, gdata) takes a network data structure net together wolffd@0: the data contribution to the error gradient wolffd@0: for a set of inputs and targets. wolffd@0: It returns the regularised error gradient using any zero mean Gaussian priors wolffd@0: on the weights defined in wolffd@0: net. In addition, if a mask is defined in net, then wolffd@0: the entries in g that correspond to weights with a 0 in the wolffd@0: mask are removed. wolffd@0: wolffd@0:

[g, gdata, gprior] = gbayes(net, gdata) additionally returns the wolffd@0: data and prior components of the error. wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: errbayes, glmgrad, mlpgrad, rbfgrad

wolffd@0: Pages: wolffd@0: Index wolffd@0:

wolffd@0: