← Back to branch summary

~ubuntu-branches/ubuntu/hoary/scilab/hoary

« back to all changes in this revision

Viewing changes to man/fr/nonlinear/numdiff.xml

  • Committer: Bazaar Package Importer
  • Author(s): Torsten Werner
  • Date: 2005-01-09 22:58:21 UTC
  • mfrom: (1.1.1 upstream)
  • Revision ID: james.westby@ubuntu.com-20050109225821-473xr8vhgugxxx5j
Tags: 3.0-12
http://bugs.debian.org/255869
changed configure.in to build scilab's own malloc.o, closes: #255869

Show diffs side-by-side

added added

removed removed

Lines of Context:
man/fr/nonlinear/numdiff.xml
 
1
<?xml version="1.0" encoding="ISO-8859-1" standalone="no"?>
 
2
<!DOCTYPE MAN SYSTEM "../../manrev.dtd">
 
3
<MAN>
 
4
  <LANGUAGE>eng</LANGUAGE>
 
5
  <TITLE>numdiff</TITLE>
 
6
  <TYPE>Scilab Function</TYPE>
 
7
  <DATE>August 1993</DATE>
 
8
  <SHORT_DESCRIPTION name="numdiff"> numerical gradient estimation</SHORT_DESCRIPTION>
 
9
  <CALLING_SEQUENCE>
 
10
    <CALLING_SEQUENCE_ITEM>g=numdiff(fun,x [,dx])  </CALLING_SEQUENCE_ITEM>
 
11
  </CALLING_SEQUENCE>
 
12
  <PARAM>
 
13
    <PARAM_INDENT>
 
14
      <PARAM_ITEM>
 
15
        <PARAM_NAME>fun</PARAM_NAME>
 
16
        <PARAM_DESCRIPTION>
 
17
          <SP>: an external, Scilab function or list. See below for calling sequence, see also external for details about external functions.</SP>
 
18
        </PARAM_DESCRIPTION>
 
19
      </PARAM_ITEM>
 
20
      <PARAM_ITEM>
 
21
        <PARAM_NAME>x</PARAM_NAME>
 
22
        <PARAM_DESCRIPTION>
 
23
          <SP>: vector, the argument of the function  <VERB>fun</VERB></SP>
 
24
        </PARAM_DESCRIPTION>
 
25
      </PARAM_ITEM>
 
26
      <PARAM_ITEM>
 
27
        <PARAM_NAME>dx</PARAM_NAME>
 
28
        <PARAM_DESCRIPTION>
 
29
          <SP>: vector, the finite difference step. Default value is  <VERB>dx=sqrt(%eps)*(1+1d-3*abs(x))</VERB></SP>
 
30
        </PARAM_DESCRIPTION>
 
31
      </PARAM_ITEM>
 
32
      <PARAM_ITEM>
 
33
        <PARAM_NAME>g</PARAM_NAME>
 
34
        <PARAM_DESCRIPTION>
 
35
          <SP>: vector, the estimated gradient</SP>
 
36
        </PARAM_DESCRIPTION>
 
37
      </PARAM_ITEM>
 
38
    </PARAM_INDENT>
 
39
  </PARAM>
 
40
  <DESCRIPTION>
 
41
    <P>
 
42
    given a function <VERB>fun(x)</VERB> from R^n to R^p
 
43
    computes the matrix <VERB>g</VERB> such as</P>
 
44
    <VERBATIM>
 
45
<![CDATA[
 
46
    [   d f    ]
 
47
    [      i   ]
 
48
g = [   ----   ]
 
49
 ij [   d x    ]
 
50
    [      j   ]
 
51
   ]]>
 
52
    </VERBATIM>
 
53
    <P>
 
54
     using finite difference methods.</P>
 
55
    <P>
 
56
    The function fun calling sequence must be <VERB>y=fun(x,p1,p2,..pn)</VERB>.If
 
57
    parameters <VERB>p1,p2,..pn</VERB> exist then <VERB>numdiff</VERB> can be called as
 
58
    follow g=numdiff(list(fun,p1,p2,..pn),x).</P>
 
59
  </DESCRIPTION>
 
60
  <EXAMPLE>
 
61
<![CDATA[
 
62
function  f=myfun(x,y,z,t)
 
63
f=(x+y)^t+z
 
64
endfunction
 
65
 
 
66
y=3;z=4;t=2;
 
67
g=numdiff(list(myfun,3,4,2),1)
 
68
 
 
69
x=1;
 
70
exact=t*(x+y)^(t-1)
 
71
 ]]>
 
72
  </EXAMPLE>
 
73
  <SEE_ALSO>
 
74
    <SEE_ALSO_ITEM>
 
75
      <LINK>optim</LINK>
 
76
    </SEE_ALSO_ITEM>
 
77
  </SEE_ALSO>
 
78
</MAN>