-
Notifications
You must be signed in to change notification settings - Fork 1
/
optimization_gradient_fixedstep.m
98 lines (90 loc) · 5.28 KB
/
optimization_gradient_fixedstep.m
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
%*******************************************************************************
%This Matlab function "grad_fixed_step" implements the gradient descent method
% using a fixed step size.
% *
%*****************************************************************************0*
% %**************************
% % Input Parameters *
% %**************************
%
% obj_f % The objective function to minimize %
% starting_pt % The starting point %
% grad_f % The gradient of the objective function %
% alpha_0 % The initial step-size %
% k_max % The maximum number of iterations %
% eps_tol % The stopping convergence tolerance %
%
%
% %**************************
% % Output Parameters %
% %**************************
%
% x_opt % The optimal solution %
% f_opt % obj_f(x_opt) %
% g_opt % grad_f(x_opt) %
% status % An integer that indicates the termination
% status of the method. I.e,
% status =1 : the method converged
% statut =-1 : the maximum number of iterations is reached
% nit % The number of iterations performed by the method
% f_count % The number of evaluation of obj_f %
% g_count % The number of evaluation of grad_f %
%
% Contact: Y. Diouane ([email protected]) --
% (C) Institut Sup�rieur de l'A�ronautique et de l'Espace (ISAE-Supa�ro)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [ ...
x_opt , ...
f_opt , ...
status , ...
nit ...
] ...
= grad_fixed_step( ...
obj_f , ...
starting_pt , ...
grad_f , ...
alpha_0 , ...
k_max , ...
eps_tol ...
)
%**********************
% CODE *
%**********************
global f_count ; % The number of evaluation of obj_f
global g_count ; % The number of evaluation of grad_f
global h_count ; % The number of evaluation of Hess_f
f_count = 0 ;
g_count = 0 ;
h_count = 0 ;
% Initialization
x_k = starting_pt ;
n = length(x_k ) ;
k = 0 ;
status = 0 ;
while(status==0)
old_x_k = x_k ;
%
% TO DO - COMPLETE THE UPDATE OF x_k
d_k = - grad_f(x_k);
alpha_k = alpha_0;
x_k = old_x_k + alpha_k*d_k;
k = k + 1 ;
%plot the current point
plot([old_x_k(1) x_k(1) ], [old_x_k(2) x_k(2)],'ko-') ;
refresh
%
% TO DO - INCLUDE A STOPPING CRITERION
%
if ( norm( grad_f(x_k) )<=eps_tol )
status= -1 ;
end
if ( k==k_max )
status= -1 ;
end
end
x_opt = x_k ;
f_opt = feval(obj_f,x_opt) ;
nit = k ;
%plot the final point
plot(x_opt(1),x_opt(2),'b*','markersize',20)
return ;