?? rbf.htm
字號:
<html><head><title>Netlab Reference Manual rbf</title></head><body><H1> rbf</H1><h2>Purpose</h2>Creates an RBF network with specified architecture<p><h2>Synopsis</h2><PRE>net = rbf(nin, nhidden, nout, rbfunc)net = rbf(nin, nhidden, nout, rbfunc, outfunc)net = rbf(nin, nhidden, nout, rbfunc, outfunc, prior, beta)</PRE><p><h2>Description</h2><CODE>net = rbf(nin, nhidden, nout, rbfunc)</CODE> constructs and initialisesa radial basis function network returning a data structure <CODE>net</CODE>.The weights are all initialised with a zero mean, unit variance normaldistribution, with the exception of the variances, which are set to one.This makes use of the Matlab function<CODE>randn</CODE> and so the seed for the random weight initialization can be set using <CODE>randn('state', s)</CODE> where <CODE>s</CODE> is the seed value. Theactivation functions are defined in terms of the distance betweenthe data point and the corresponding centre. Note that the functions arecomputed to a convenient constant multiple: for example, the Gaussianis not normalised. (Normalisation is not needed as the function outputsare linearly combined in the next layer.)<p>The fields in <CODE>net</CODE> are<PRE> type = 'rbf' nin = number of inputs nhidden = number of hidden units nout = number of outputs nwts = total number of weights and biases actfn = string defining hidden unit activation function: 'gaussian' for a radially symmetric Gaussian function. 'tps' for r^2 log r, the thin plate spline function. 'r4logr' for r^4 log r. outfn = string defining output error function: 'linear' for linear outputs (default) and SoS error. 'neuroscale' for Sammon stress measure. c = centres wi = squared widths (null for rlogr and tps) w2 = second layer weight matrix b2 = second layer bias vector</PRE><p><CODE>net = rbf(nin, nhidden, nout, rbfund, outfunc)</CODE> allows the user tospecify the type of error function to be used. The field <CODE>outfn</CODE>is set to the value of this string. Linear outputs (for regression problems)and Neuroscale outputs (for topographic mappings) are supported.<p><CODE>net = rbf(nin, nhidden, nout, rbfunc, outfunc, prior, beta)</CODE>,in which <CODE>prior</CODE> isa scalar, allows the field <CODE>net.alpha</CODE> in the data structure<CODE>net</CODE> to be set, corresponding to a zero-mean isotropic Gaussianprior with inverse variance with value <CODE>prior</CODE>. Alternatively,<CODE>prior</CODE> can consist of a data structure with fields <CODE>alpha</CODE>and <CODE>index</CODE>, allowing individual Gaussian priors to be set overgroups of weights in the network. Here <CODE>alpha</CODE> is a column vectorin which each element corresponds to a separate group of weights,which need not be mutually exclusive. The membership of the groups isdefined by the matrix <CODE>indx</CODE> in which the columns correspond tothe elements of <CODE>alpha</CODE>. Each column has one element for eachweight in the matrix, in the order defined by the function<CODE>rbfpak</CODE>, and each element is 1 or 0 according to whether theweight is a member of the corresponding group or not. A utilityfunction <CODE>rbfprior</CODE> is provided to help in setting up the<CODE>prior</CODE> data structure.<p><CODE>net = rbf(nin, nhidden, nout, func, prior, beta)</CODE> also sets the additional field <CODE>net.beta</CODE> in the data structure <CODE>net</CODE>, wherebeta corresponds to the inverse noise variance.<p><h2>Example</h2>The following code constructs an RBF network with 1 input and output nodeand 5 hidden nodes and then propagates some data <CODE>x</CODE> through it.<PRE>net = rbf(1, 5, 1, 'tps');[y, act] = rbffwd(net, x);</PRE><p><h2>See Also</h2><CODE><a href="rbferr.htm">rbferr</a></CODE>, <CODE><a href="rbffwd.htm">rbffwd</a></CODE>, <CODE><a href="rbfgrad.htm">rbfgrad</a></CODE>, <CODE><a href="rbfpak.htm">rbfpak</a></CODE>, <CODE><a href="rbftrain.htm">rbftrain</a></CODE>, <CODE><a href="rbfunpak.htm">rbfunpak</a></CODE><hr><b>Pages:</b><a href="index.htm">Index</a><hr><p>Copyright (c) Ian T Nabney (1996-9)</body></html>?? 快捷鍵說明
復制代碼
Ctrl + C
搜索代碼
Ctrl + F
全屏模式
F11
切換主題
Ctrl + Shift + D
顯示快捷鍵
?
增大字號
Ctrl + =
減小字號
Ctrl + -