annotate toolboxes/FullBNT-1.0.7/nethelp3.3/netinit.htm @ 0:cc4b1211e677 tip

initial commit to HG from Changeset: 646 (e263d8a21543) added further path and more save "camirversion.m"
author Daniel Wolff
date Fri, 19 Aug 2016 13:07:06 +0200
parents
children
rev   line source
Daniel@0 1 <html>
Daniel@0 2 <head>
Daniel@0 3 <title>
Daniel@0 4 Netlab Reference Manual netinit
Daniel@0 5 </title>
Daniel@0 6 </head>
Daniel@0 7 <body>
Daniel@0 8 <H1> netinit
Daniel@0 9 </H1>
Daniel@0 10 <h2>
Daniel@0 11 Purpose
Daniel@0 12 </h2>
Daniel@0 13 Initialise the weights in a network.
Daniel@0 14
Daniel@0 15 <p><h2>
Daniel@0 16 Synopsis
Daniel@0 17 </h2>
Daniel@0 18 <PRE>
Daniel@0 19 net = netinit(net, prior)
Daniel@0 20 </PRE>
Daniel@0 21
Daniel@0 22
Daniel@0 23 <p><h2>
Daniel@0 24 Description
Daniel@0 25 </h2>
Daniel@0 26
Daniel@0 27 <p><CODE>net = netinit(net, prior)</CODE> takes a network data structure
Daniel@0 28 <CODE>net</CODE> and sets the weights and biases by sampling from a Gaussian
Daniel@0 29 distribution. If <CODE>prior</CODE> is a scalar, then all of the parameters
Daniel@0 30 (weights and biases) are sampled from a single isotropic Gaussian with
Daniel@0 31 inverse variance equal to <CODE>prior</CODE>. If <CODE>prior</CODE> is a data
Daniel@0 32 structure of the kind generated by <CODE>mlpprior</CODE>, then the parameters
Daniel@0 33 are sampled from multiple Gaussians according to their groupings
Daniel@0 34 (defined by the <CODE>index</CODE> field) with corresponding variances
Daniel@0 35 (defined by the <CODE>alpha</CODE> field).
Daniel@0 36
Daniel@0 37 <p><h2>
Daniel@0 38 See Also
Daniel@0 39 </h2>
Daniel@0 40 <CODE><a href="mlpprior.htm">mlpprior</a></CODE>, <CODE><a href="netunpak.htm">netunpak</a></CODE>, <CODE><a href="rbfprior.htm">rbfprior</a></CODE><hr>
Daniel@0 41 <b>Pages:</b>
Daniel@0 42 <a href="index.htm">Index</a>
Daniel@0 43 <hr>
Daniel@0 44 <p>Copyright (c) Ian T Nabney (1996-9)
Daniel@0 45
Daniel@0 46
Daniel@0 47 </body>
Daniel@0 48 </html>