-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathLISWET5.SIF
211 lines (138 loc) · 4.61 KB
/
LISWET5.SIF
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
***************************
* SET UP THE INITIAL DATA *
***************************
NAME LISWET5
* Problem :
* *********
* A k-convex approximation problem posed as a
* convex quadratic problem, with variable dimensions.
* Formulation:
* -----------
* n+k 2
* minimize 1/2 sum ( x - c )
* i=1 i i
* subject to
* k k-i
* sum ( k ) ( -1 ) x > 0
* i=0 ( i ) j+i =
* where c = g( t ) + small perturbation, t = (i-1)/(n+k-1)
* i i i
* Case 5: g(t) = exp(t)
* NB. Perturbations are not random as Li and Swetits's
* random number generator is undefined.
* Source:
* W. Li and J. Swetits,
* "A Newton method for convex regression, data smoothing and
* quadratic programming with bounded constraints",
* SIAM J. Optimization 3 (3) pp 466-488, 1993.
* SIF input: Nick Gould, August 1994.
* classification QLR2-AN-V-V
*IE N 100 $-PARAMETER 103 variables original value
*IE K 3 $-PARAMETER original value
*IE N 100 $-PARAMETER 104 variables
*IE K 4 $-PARAMETER
*IE N 100 $-PARAMETER 105 variables
*IE K 5 $-PARAMETER
*IE N 100 $-PARAMETER 106 variables
*IE K 6 $-PARAMETER
*IE N 400 $-PARAMETER 402 variables
*IE K 2 $-PARAMETER
*IE N 400 $-PARAMETER 403 variables
*IE K 3 $-PARAMETER
*IE N 2000 $-PARAMETER 2001 variables
*IE K 1 $-PARAMETER
IE N 2000 $-PARAMETER 2002 variables
IE K 2 $-PARAMETER
*IE N 10000 $-PARAMETER 10001 variables
*IE K 1 $-PARAMETER
*IE N 10000 $-PARAMETER 10002 variables
*IE K 2 $-PARAMETER
* Constants
IE 0 0
IE 1 1
RE ONE 1.0
RE HALF 0.5
* Set some useful parameters
I+ N+K N K
IA N+K-1 N+K -1
RI RN+K-1 N+K-1
RE CONST 0.0
* Binomial coefficients
A= B(0) ONE
DO I 1 K
IA I-1 I -1
RI RI I
A* B(I) B(I-1) RI
ND
* (-1)^K * Binomial coefficients
A= C(0) ONE
R= PLUSMINUS ONE
DO I 1 K
I- K-I K I
RM PLUSMINUS PLUSMINUS -1.0
A/ C(I) B(K) B(I)
A/ C(I) C(I) B(K-I)
A* C(I) C(I) PLUSMINUS
ND
VARIABLES
DO I 1 N+K
X X(I)
ND
GROUPS
* Objective linear coefficients
DO I 1 N+K
IA I-1 I -1
RI RI I
RI RI-1 I-1
R/ TI RI-1 RN+K-1
R( GT EXP TI
* Perturb by 0.1 * sin( i )
R( RANDOM SIN RI
RM RANDOM RANDOM 0.1
R+ CI GT RANDOM
RM -CI CI -1.0
R* -CI*CI -CI CI
R+ CONST CONST -CI*CI
ZN OBJ X(I) -CI
ND
* Divided difference constraints
DO J 1 N
I+ J+K J K
DO I 0 K
I- J+K-I J+K I
ZG CON(J) X(J+K-I) C(I)
OD I
OD J
CONSTANTS
R* CONST HALF CONST
Z LISWET5 OBJ CONST
BOUNDS
FR LISWET5 'DEFAULT'
ELEMENT TYPE
EV SQ X
ELEMENT USES
* The elements corresponding to the squre variables
DO I 1 N+K
XT XSQ(I) SQ
ZV XSQ(I) X X(I)
ND
GROUP USES
* The diagonal elements
DO I 1 N+K
XE OBJ XSQ(I)
OD I
OBJECT BOUND
* Solution
*LO SOLTN
ENDATA
***********************
* SET UP THE FUNCTION *
* AND RANGE ROUTINES *
***********************
ELEMENTS LISWET5
INDIVIDUALS
T SQ
F 5.0D-1 * X * X
G X X
H X X 1.0D+0
ENDATA