Mercurial > hg > camir-aes2014

comparison toolboxes/FullBNT-1.0.7/nethelp3.3/glm.htm @ 0:e9a9cd732c1e tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:e9a9cd732c1e
1 <html>
2 <head>
3 <title>
4 Netlab Reference Manual glm
5 </title>
6 </head>
7 <body>
8 <H1> glm
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Create a generalized linear model.
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19 net = glm(nin, nout, func)
20 net = glm(nin, nout, func, prior)
21 net = glm(nin, nout, func, prior, beta)
22 </PRE>
23
24
25 <p><h2>
26 Description
27 </h2>
28
29 <p><CODE>net = glm(nin, nout, func)</CODE> takes the number of inputs
30 and outputs for a generalized linear model, together
31 with a string <CODE>func</CODE> which specifies the output unit activation function,
32 and returns a data structure <CODE>net</CODE>. The weights are drawn from a zero mean,
33 isotropic Gaussian, with variance scaled by the fan-in of the
34 output units. This makes use of the Matlab function
35 <CODE>randn</CODE> and so the seed for the random weight initialization can be
36 set using <CODE>randn('state', s)</CODE> where <CODE>s</CODE> is the seed value. The optional
37 argument <CODE>alpha</CODE> sets the inverse variance for the weight
38 initialization.
39
40 <p>The fields in <CODE>net</CODE> are
41 <PRE>
42 type = 'glm'
43 nin = number of inputs
44 nout = number of outputs
45 nwts = total number of weights and biases
46 actfn = string describing the output unit activation function:
47 'linear'
48 'logistic'
49 'softmax'
50 w1 = first-layer weight matrix
51 b1 = first-layer bias vector
52 </PRE>
53
54
55 <p><CODE>net = glm(nin, nout, func, prior)</CODE>, in which <CODE>prior</CODE> is
56 a scalar, allows the field
57 <CODE>net.alpha</CODE> in the data structure <CODE>net</CODE> to be set, corresponding
58 to a zero-mean isotropic Gaussian prior with inverse variance with
59 value <CODE>prior</CODE>. Alternatively, <CODE>prior</CODE> can consist of a data
60 structure with fields <CODE>alpha</CODE> and <CODE>index</CODE>, allowing individual
61 Gaussian priors to be set over groups of weights in the network. Here
62 <CODE>alpha</CODE> is a column vector in which each element corresponds to a
63 separate group of weights, which need not be mutually exclusive. The
64 membership of the groups is defined by the matrix <CODE>index</CODE> in which
65 the columns correspond to the elements of <CODE>alpha</CODE>. Each column has
66 one element for each weight in the matrix, in the order defined by the
67 function <CODE>glmpak</CODE>, and each element is 1 or 0 according to whether
68 the weight is a member of the corresponding group or not.
69
70 <p><CODE>net = glm(nin, nout, func, prior, beta)</CODE> also sets the
71 additional field <CODE>net.beta</CODE> in the data structure <CODE>net</CODE>, where
72 beta corresponds to the inverse noise variance.
73
74 <p><h2>
75 See Also
76 </h2>
77 <CODE><a href="glmpak.htm">glmpak</a></CODE>, <CODE><a href="glmunpak.htm">glmunpak</a></CODE>, <CODE><a href="glmfwd.htm">glmfwd</a></CODE>, <CODE><a href="glmerr.htm">glmerr</a></CODE>, <CODE><a href="glmgrad.htm">glmgrad</a></CODE>, <CODE><a href="glmtrain.htm">glmtrain</a></CODE><hr>
78 <b>Pages:</b>
79 <a href="index.htm">Index</a>
80 <hr>
81 <p>Copyright (c) Ian T Nabney (1996-9)
82
83
84 </body>
85 </html>