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
|
% STK_BENCHMARK_PARAMESTIM A simple 1D parameter estimation benchmark
% Copyright Notice
%
% Copyright (C) 2016 CentraleSupelec
% Copyright (C) 2011-2014 SUPELEC
%
% Authors: Julien Bect <julien.bect@centralesupelec.fr>
% Emmanuel Vazquez <emmanuel.vazquez@centralesupelec.fr>
% Copying Permission Statement
%
% This file is part of
%
% STK: a Small (Matlab/Octave) Toolbox for Kriging
% (https://github.com/stk-kriging/stk/)
%
% STK is free software: you can redistribute it and/or modify it under
% the terms of the GNU General Public License as published by the Free
% Software Foundation, either version 3 of the License, or (at your
% option) any later version.
%
% STK is distributed in the hope that it will be useful, but WITHOUT
% ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
% or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
% License for more details.
%
% You should have received a copy of the GNU General Public License
% along with STK. If not, see <http://www.gnu.org/licenses/>.
function stk_benchmark_paramestim ()
NREP = 20;
for noise_std = [0 0.1],
for ni = [10 50],
t = zeros (1, NREP + 1);
for i = 1:(NREP + 1),
tic;
test_function (ni, noise_std);
t(i) = toc;
end
t = t(2:end);
fprintf ('noise_std = %.1f ', noise_std);
fprintf ('ni = %d ', ni);
t_est = median (t);
t_mad = mean (abs (t - t_est));
fprintf ('t = %.3f [%.3f]\n', t_est, t_mad);
drawnow ();
end
end
end
function test_function (ni, noise_std)
f = @(x)(- (0.8 * x + sin (5 * x + 1) + 0.1 * sin (10 * x)));
DIM = 1; % Dimension of the factor space
BOX = [-1.0; 1.0]; % Factor space
NOISY = (noise_std > 0);
NITER = 5; % number of random designs generated in stk_sampling_maximinlhs()
xi = stk_sampling_maximinlhs (ni, DIM, BOX, NITER); % evaluation points
zi = stk_feval (f, xi); % evaluation results
if NOISY,
zi = zi + noise_std * randn (ni, 1);
end
model = stk_model (@stk_materncov_iso);
if ~ NOISY,
% Noiseless case: set a small "regularization" noise
% the (log)variance of which is provided by stk_param_init
model.lognoisevariance = 1e-10;
else
% Otherwise, set the variance of the noise
% (assumed to be known, not estimated, in this example)
model.lognoisevariance = 2 * log (noise_std);
end
% Estimate the parameters
model.param = stk_param_estim (model, xi, zi, log ([1.0; 4.0; 1/0.4]));
end
|
