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
|
% STK_MODEL generates a model with default covariance parameters
%
% CALL: MODEL = stk_model ()
%
% returns a structure MODEL (see below for a description of the fields in such
% a structure) corresponding to one-dimensional Gaussian process prior with a
% constant but unknown mean ("ordinary" kriging) and a stationary Matern
% covariance function.
%
% CALL: MODEL = stk_model (COVARIANCE_TYPE)
%
% uses the user-supplied COVARIANCE_TYPE instead of the default.
%
% CALL: MODEL = stk_model (COVARIANCE_TYPE, DIM)
%
% creates a DIM-dimensional model. Note that, for DIM > 1, anisotropic
% covariance functions are provided with default parameters that make them
% isotropic.
%
% In STK, a Gaussian process model is described by a 'model' structure,
% which has mandatory fields and optional fields.
%
% MANDATORY FIELDS: covariance_type, param, lm, lognoisevariance
% OPTIONAL FIELD: param_prior, noise_prior
%
% See also stk_materncov_iso, stk_materncov_aniso, ...
% Copyright Notice
%
% Copyright (C) 2015, 2016, 2018 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 model = stk_model (covariance_type, varargin)
% Extract a handle and a name from what we are given
if nargin > 0
if isa (covariance_type, 'function_handle')
covariance_name = func2str (covariance_type);
elseif ischar (covariance_type)
covariance_name = covariance_type;
covariance_type = str2func (covariance_type);
else
stk_error (['covariance_type should be a function name '...
'or a handle to a function.'], 'TypeMismatch');
end
else
covariance_type = @stk_materncov_iso;
covariance_name = 'stk_materncov_iso';
end
if strcmp (covariance_name, 'stk_discretecov')
% special case: build a discrete model
model = stk_model_discretecov (varargin{:});
else
% general case
model = stk_model_ (covariance_type, varargin{:});
end
end % function
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% stk_model_discretecov %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function model_out = stk_model_discretecov (model_base, x)
% Make sure that lognoisevariance is -inf for noiseless models
if ~ stk_isnoisy (model_base)
model_base.lognoisevariance = -inf;
end
[K, P] = stk_make_matcov (model_base, x, x);
model_out = struct ( ...
'covariance_type', 'stk_discretecov', ...
'param', struct ('K', K, 'P', P));
if ~ isscalar (model_base.lognoisevariance)
error ('This case is not supported.');
else
model_out.lognoisevariance = model_base.lognoisevariance;
end
end % function
%%%%%%%%%%%%%%%%%%
%%% stk_model_ %%%
%%%%%%%%%%%%%%%%%%
function model = stk_model_ (covariance_type, dim)
model = struct();
model.covariance_type = covariance_type;
% use ordinary kriging as a default choice
model.lm = stk_lm_constant ();
% default dimension is d = 1
if nargin < 2
model.dim = 1;
else
model.dim = dim;
end
model.param = []; % Make it pass stk_assert_model_struct
try
model.param = stk_param_init (model);
catch
warning (lasterr ());
end
model.lognoisevariance = - inf;
end % function
%!test model = stk_model ();
%!test model = stk_model (@stk_expcov_iso);
%!test model = stk_model (@stk_expcov_iso, 1);
%!test model = stk_model (@stk_expcov_iso, 3);
%!test model = stk_model (@stk_expcov_aniso);
%!test model = stk_model (@stk_expcov_aniso, 1);
%!test model = stk_model (@stk_expcov_aniso, 3);
%!test model = stk_model (@stk_materncov_iso);
%!test model = stk_model (@stk_materncov_iso, 1);
%!test model = stk_model (@stk_materncov_iso, 3);
%!test model = stk_model (@stk_materncov_aniso);
%!test model = stk_model (@stk_materncov_aniso, 1);
%!test model = stk_model (@stk_materncov_aniso, 3);
%!test model = stk_model (@stk_materncov32_iso);
%!test model = stk_model (@stk_materncov32_iso, 1);
%!test model = stk_model (@stk_materncov32_iso, 3);
%!test model = stk_model (@stk_materncov32_aniso);
%!test model = stk_model (@stk_materncov32_aniso, 1);
%!test model = stk_model (@stk_materncov32_aniso, 3);
%!test model = stk_model (@stk_materncov52_iso);
%!test model = stk_model (@stk_materncov52_iso, 1);
%!test model = stk_model (@stk_materncov52_iso, 3);
%!test model = stk_model (@stk_materncov52_aniso);
%!test model = stk_model (@stk_materncov52_aniso, 1);
%!test model = stk_model (@stk_materncov52_aniso, 3);
%!test model = stk_model (@stk_gausscov_iso);
%!test model = stk_model (@stk_gausscov_iso, 1);
%!test model = stk_model (@stk_gausscov_iso, 3);
%!test model = stk_model (@stk_gausscov_aniso);
%!test model = stk_model (@stk_gausscov_aniso, 1);
%!test model = stk_model (@stk_gausscov_aniso, 3);
%!test model = stk_model (@stk_sphcov_iso);
%!test model = stk_model (@stk_sphcov_iso, 1);
%!test model = stk_model (@stk_sphcov_iso, 3);
%!test model = stk_model (@stk_sphcov_aniso);
%!test model = stk_model (@stk_sphcov_aniso, 1);
%!test model = stk_model (@stk_sphcov_aniso, 3);
|
