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
|
% STK_PREDICT_LEAVEONEOUT_DIRECT [STK internal]
% Copyright Notice
%
% Copyright (C) 2016-2018, 2020 CentraleSupelec
%
% Author: Julien Bect <julien.bect@centralesupelec.fr>
% Stefano Duhamel <stefano.duhamel@supelec.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 [LOO_pred, LOO_res] = stk_predict_leaveoneout_direct (M_post)
% Heteroscedatic noise ?
heteroscedastic = ~ isscalar (M_post.prior_model.lognoisevariance);
n = stk_get_sample_size (M_post);
zp_mean = zeros (n, 1);
zp_var = zeros (n, 1);
prior_model = M_post.prior_model;
for i = 1:n
xx = M_post.input_data; xx(i, :) = []; xt = M_post.input_data(i, :);
zz = M_post.output_data; zz(i, :) = [];
% In the heteroscedastic case, the vector of log-variances for the
% noise is stored in prior_model.lognoisevariance. This vector must be
% modified too, when performing cross-validation.
if heteroscedastic
prior_model = M_post.prior_model;
prior_model.lognoisevariance(i) = [];
end
zp = stk_predict (prior_model, xx, zz, xt);
zp_mean(i) = zp.mean;
zp_var(i) = zp.var;
end
% Prepare outputs
LOO_pred = stk_dataframe ([zp_mean zp_var], {'mean', 'var'});
% Compute residuals ?
if nargout ~= 1
% Compute "raw" residuals
raw_res = M_post.output_data - zp_mean;
% Compute normalized residual
noisevariance = stk_get_observation_variances (M_post);
norm_res = raw_res ./ (sqrt (noisevariance + zp_var));
% Pack results into a dataframe
LOO_res = stk_dataframe ([raw_res norm_res], ...
{'residuals', 'norm_res'});
end
% Create LOO cross-validation plots?
if nargout == 0
% Plot predictions VS observations (left planel)...
stk_subplot (1, 2, 1); stk_plot_predvsobs (M_post.output_data, LOO_pred);
% ...and normalized residuals (right panel)
stk_subplot (1, 2, 2); stk_plot_histnormres (LOO_res.norm_res);
end
end % function
|
