1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
|
<HTML><HEAD><TITLE>Manpage of HYBRD_</TITLE>
</HEAD><BODY>
<H1>HYBRD_</H1>
Section: C Library Functions (3)<BR>Updated: March 8, 2002<BR><A HREF="#index">Index</A>
<A HREF="index.html">Return to Main Contents</A><HR>
<A NAME="lbAB"> </A>
<H2>NAME</H2>
hybrd_, hybrd1_ - find a zero of a system of nonlinear function
<A NAME="lbAC"> </A>
<H2>SYNOPSIS</H2>
<B>#include <<A HREF="file:/usr/include/minpack.h">minpack.h</A>></B>
<P>
<DL COMPACT>
<DT>
<B>void hybrd1_ ( </B>
<B>void (*</B><I>fcn</I><B>)(</B>
<B>int *</B><I>n</I><B>,</B>
<B>double *</B><I>x</I><B>,</B>
<DD>
<B>double *</B><I>fvec</I><B>,</B>
<B>int *</B><I>iflag</I><B>),</B>
<DL COMPACT><DT><DD>
<B>int *</B><I>n</I><B>,</B>
<B>double *</B><I>x</I><B>,</B>
<B>double *</B><I>fvec</I><B>,</B>
<DD>
<B>double *</B><I>tol</I><B>,</B>
<B>int *</B><I>info</I><B>,</B>
<B>double *</B><I>wa</I><B>,</B>
<DD>
<B>int *</B><I>lwa</I><B>);</B>
</DL>
<DT>
<B>void hybrd_</B>
<B>( void (*</B><I>fcn</I><B>)(</B>
<B>int * </B><I>n</I><B>,</B>
<B>double *</B><I>x</I><B>,</B>
<DD>
<B>double *</B><I>fvec</I><B>,</B>
<B>int *</B><I>iflag</I><B>),</B>
<DL COMPACT><DT><DD>
<B>int *</B><I>n</I><B>,</B>
<B>double *</B><I>x</I><B>,</B>
<B>double *</B><I>fvec</I><B>,</B>
<DD>
<B>double *</B><I>xtol</I><B>,</B>
<B>int *</B><I>maxfev</I><B>,</B>
<B>int *</B><I>ml</I><B>,</B>
<B>int *</B><I>mu</I><B>,</B>
<DD>
<B>double *</B><I>epsfcn</I><B>,</B>
<B>double *</B><I>diag</I><B>,</B>
<B>int *</B><I>mode</I><B>,</B>
<B>double *</B><I>factor</I><B>,</B>
<B>int *</B><I>nprint</I><B>,</B>
<B>int *</B><I>info</I><B>,</B>
<DD>
<B>int *</B><I>nfev</I><B>,</B>
<B>double *</B><I>fjac</I><B>,</B>
<B>int *</B><I>ldfjac</I><B>,</B>
<DD>
<B>double *</B><I>r</I><B>,</B>
<B>int *</B><I>lr</I><B>,</B>
<B>double *</B><I>qtf</I><B>,</B>
<DD>
<B>double *</B><I>wa1</I><B>,</B>
<B>double *</B><I>wa2</I><B>,</B>
<B>double *</B><I>wa3</I><B>,</B>
<B>double *</B><I>wa4</I><B>);</B>
</DL>
<DD>
</DL>
<A NAME="lbAD"> </A>
<H2>DESCRIPTION</H2>
<DD>The purpose of <B>hybrd_</B> is to find a zero of a system of
<I>n</I> nonlinear functions in <I>n</I> variables by a modification
of the Powell hybrid method. The user must provide a
subroutine which calculates the functions. The Jacobian is
then calculated by a forward-difference approximation.
<P>
<B>hybrd1_</B> serves the same function but has a simplified calling
sequence.
<BR>
<A NAME="lbAE"> </A>
<H3>Language notes</H3>
<B>hybrd_</B> and <B>hybrd1_</B> are written in FORTRAN. If calling from
C, keep these points in mind:
<DL COMPACT>
<DT>Name mangling.<DD>
With <B>g77</B> version 2.95 or 3.0, all the function names end in an
underscore. This may change with future versions of <B>g77</B>.
<DT>Compile with <B>g77</B>.<DD>
Even if your program is all C code, you should link with <B>g77</B>
so it will pull in the FORTRAN libraries automatically. It's easiest
just to use <B>g77</B> to do all the compiling. (It handles C just fine.)
<DT>Call by reference.<DD>
All function parameters must be pointers.
<DT>Column-major arrays.<DD>
Suppose a function returns an array with 5 rows and 3 columns in an
array <I>z</I> and in the call you have declared a leading dimension of
7. The FORTRAN and equivalent C references are:
<P>
<PRE>
z(1,1) z[0]
z(2,1) z[1]
z(5,1) z[4]
z(1,2) z[7]
z(1,3) z[14]
z(i,j) z[(i-1) + (j-1)*7]
</PRE>
</DL>
<A NAME="lbAF"> </A>
<H3>Parameters for both functions</H3>
<I>fcn</I> is the name of the user-supplied subroutine which calculates
the functions. In FORTRAN, <I>fcn</I> must be declared in an external
statement in the user calling program, and should be written as
follows:
<P>
<PRE>
subroutine fcn(n,x,fvec,iflag)
integer n,iflag
double precision x(n),fvec(n)
----------
calculate the functions at x and
return this vector in fvec.
---------
return
end
</PRE>
<P>
<P>
In C, <I>fcn</I> should be written as follows:
<P>
<PRE>
void fcn(int n, double *x, double *fvec, int *iflag)
{
/* calculate the functions at x and
return this vector in fvec. */
}
</PRE>
<P>
The value of <I>iflag</I> should not be changed by <I>fcn</I> unless
the user wants to terminate execution of <B>hybrd_</B>.
In this case set <I>iflag</I> to a negative integer.
<P>
<I>n</I> is a positive integer input variable set to the number
of functions and variables.
<P>
<I>x</I> is an array of length <I>n</I>. On input <I>x</I> must contain
an initial estimate of the solution vector. On output <I>x</I>
contains the final estimate of the solution vector.
<P>
<I>fvec</I> is an output array of length <I>n</I> which contains
the functions evaluated at the output <I>x</I>.
<BR>
<A NAME="lbAG"> </A>
<H3>Parameters for <B>hybrd1_</B></H3>
<P>
<I>tol</I> is a nonnegative input variable. Termination occurs
when the algorithm estimates that the relative error
between <I>x</I> and the solution is at most <I>tol</I>.
<P>
<I>info</I> is an integer output variable. If the user has
terminated execution, <I>info</I> is set to the (negative)
value of <I>iflag</I>. See description of <I>fcn</I>. Otherwise,
<I>info</I> is set as follows.
<P>
<I>info</I> = 0 improper input parameters.
<P>
<I>info</I> = 1 algorithm estimates that the relative error
<BR> between <I>x</I> and the solution is at most <I>tol</I>.
<P>
<I>info</I> = 2 number of calls to fcn has reached or exceeded
<BR> 200*(<I>n</I>+1).
<P>
<I>info</I> = 3 <I>tol</I> is too small. No further improvement in
<BR> the approximate solution <I>x</I> is possible.
<P>
<I>info</I> = 4 iteration is not making good progress.
<P>
<I>wa</I> is a work array of length <I>lwa</I>.
<P>
<I>lwa</I> is a positive integer input variable not less than
(<I>n</I>*(3*<I>n</I>+13))/2.
<BR>
<A NAME="lbAH"> </A>
<H3>Parameters for <B>hybrd_</B></H3>
<P>
<I>xtol</I> is a nonnegative input variable. Termination
occurs when the relative error between two consecutive
iterates is at most <I>xtol</I>.
<P>
<I>maxfev</I> is a positive integer input variable. Termination
occurs when the number of calls to <I>fcn</I> is at least <I>maxfev</I>
by the end of an iteration.
<P>
<I>ml</I> is a nonnegative integer input variable which specifies
the number of subdiagonals within the band of the
jacobian matrix. If the Jacobian is not banded, set
<I>ml</I> to at least <I>n</I> - 1.
<P>
<I>mu</I> is a nonnegative integer input variable which specifies
the number of superdiagonals within the band of the
jacobian matrix. If the jacobian is not banded, set
mu to at least <I>n</I> - 1.
<P>
<I>epsfcn</I> is an input variable used in determining a suitable
step length for the forward-difference approximation. This
approximation assumes that the relative errors in the
functions are of the order of <I>epsfcn</I>. If <I>epsfcn</I> is less
than the machine precision, it is assumed that the relative
errors in the functions are of the order of the machine
precision.
<P>
<I>diag</I> is an array of length <I>n</I>. If <I>mode</I> = 1 (see
below), <I>diag</I> is internally set. If <I>mode</I> = 2, <I>diag</I>
must contain positive entries that serve as
multiplicative scale factors for the variables.
<P>
<I>mode</I> is an integer input variable. If <I>mode</I> = 1, the
variables will be scaled internally. If <I>mode</I> = 2,
the scaling is specified by the input <I>diag</I>. Other
values of mode are equivalent to <I>mode</I> = 1.
<P>
<I>factor</I> is a positive input variable used in determining the
initial step bound. This bound is set to the product of
<I>factor</I> and the euclidean norm of diag*x if nonzero, or else
to <I>factor</I> itself. In most cases factor should lie in the
interval (.1,100.). 100. Is a generally recommended value.
<P>
<I>nprint</I> is an integer input variable that enables controlled
printing of iterates if it is positive. In this case,
<I>fcn</I> is called with <I>iflag</I> = 0 at the beginning of the first
iteration and every nprint iterations thereafter and
immediately prior to return, with <I>x</I> and <I>fvec</I> available
for printing. If <I>nprint</I> is not positive, no special calls
of <I>fcn</I> with <I>iflag</I> = 0 are made.
<P>
<I>info</I> is an integer output variable. If the user has
terminated execution, <I>info</I> is set to the (negative)
value of <I>iflag</I>. See description of <I>fcn</I>. Otherwise,
<I>info</I> is set as follows.
<P>
<I>info</I> = 0 improper input parameters.
<P>
<I>info</I> = 1 relative error between two consecutive iterates
<BR> is at most <I>xtol</I>.
<P>
<I>info</I> = 2 number of calls to <I>fcn</I> has reached or exceeded
<BR> <I>maxfev</I>.
<P>
<I>info</I> = 3 <I>xtol</I> is too small. No further improvement in
<BR> the approximate solution <I>x</I> is possible.
<P>
<I>info</I> = 4 iteration is not making good progress, as
<BR> measured by the improvement from the last
<BR> five jacobian evaluations.
<P>
<I>info</I> = 5 iteration is not making good progress, as
<BR> measured by the improvement from the last
<BR> ten iterations.
<P>
<I>nfev</I> is an integer output variable set to the number of
calls to <I>fcn</I>.
<P>
<I>fjac</I> is an output <I>n</I> by <I>n</I> array which contains the
orthogonal matrix <I>q</I> produced by the <I>qr</I> factorization
of the final approximate jacobian.
<P>
<I>ldfjac</I> is a positive integer input variable not less than <I>n</I>
which specifies the leading dimension of the array <I>fjac</I>.
<P>
<I>r</I> is an output array of length <I>lr</I> which contains the
upper triangular matrix produced by the <I>qr</I> factorization
of the final approximate Jacobian, stored rowwise.
<P>
<I>lr</I> is a positive integer input variable not less than
(<I>n</I>*(<I>n</I>+1))/2.
<P>
<I>qtf</I> is an output array of length <I>n</I> which contains
the vector (q transpose)*<I>fvec</I>.
<P>
<I>wa1</I>, <I>wa2</I>, <I>wa3</I>, and <I>wa4</I> are work arrays of length <I>n</I>.
<P>
<A NAME="lbAI"> </A>
<H2>SEE ALSO</H2>
<B><A HREF="hybrj_.html">hybrj</A></B>(3),
<B><A HREF="hybrj_.html">hybrj1</A></B>(3).
<BR>
<P>
<A NAME="lbAJ"> </A>
<H2>AUTHORS</H2>
Burton S. Garbow, Kenneth E. Hillstrom, Jorge J. More.
<BR>
This manual page was written by Jim Van Zandt <<A HREF="mailto:jrv@debian.org">jrv@debian.org</A>>,
for the Debian GNU/Linux system (but may be used by others).
<P>
<HR>
<A NAME="index"> </A><H2>Index</H2>
<DL>
<DT><A HREF="#lbAB">NAME</A><DD>
<DT><A HREF="#lbAC">SYNOPSIS</A><DD>
<DT><A HREF="#lbAD">DESCRIPTION</A><DD>
<DL>
<DT><A HREF="#lbAE">Language notes</A><DD>
<DT><A HREF="#lbAF">Parameters for both functions</A><DD>
<DT><A HREF="#lbAG">Parameters for <B>hybrd1_</B></A><DD>
<DT><A HREF="#lbAH">Parameters for <B>hybrd_</B></A><DD>
</DL>
<DT><A HREF="#lbAI">SEE ALSO</A><DD>
<DT><A HREF="#lbAJ">AUTHORS</A><DD>
</DL>
<HR>
This document was created by
<A HREF="index.html">man2html</A>,
using the manual pages.<BR>
Time: 10:19:50 GMT, April 20, 2007
</BODY>
</HTML>
|
