/
SingleFITCLaplacianInferenceMethodWithLBFGS_unittest.cc
458 lines (374 loc) · 14.9 KB
/
SingleFITCLaplacianInferenceMethodWithLBFGS_unittest.cc
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
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
/*
* Copyright (c) The Shogun Machine Learning Toolbox
* Written (W) 2015 Wu Lin
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
* ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* The views and conclusions contained in the software and documentation are those
* of the authors and should not be interpreted as representing official policies,
* either expressed or implied, of the Shogun Development Team.
*
*/
#include <shogun/lib/config.h>
#if defined(HAVE_EIGEN3) && defined(HAVE_LINALG_LIB)
#include <shogun/labels/BinaryLabels.h>
#include <shogun/features/DenseFeatures.h>
#include <shogun/kernel/GaussianKernel.h>
#include <shogun/machine/gp/ConstMean.h>
#include <shogun/machine/gp/ZeroMean.h>
#include <shogun/machine/gp/GaussianARDFITCKernel.h>
#include <shogun/machine/gp/SingleFITCLaplacianInferenceMethodWithLBFGS.h>
#include <shogun/machine/gp/LogitLikelihood.h>
#include <shogun/mathematics/Math.h>
#include <gtest/gtest.h>
using namespace shogun;
TEST(SingleFITCLaplacianInferenceMethodWithLBFGS,get_cholesky)
{
index_t n=6;
index_t dim=2;
index_t m=3;
float64_t rel_tolorance=1e-5;
float64_t abs_tolorance;
SGMatrix<float64_t> feat_train(dim, n);
SGMatrix<float64_t> lat_feat_train(dim, m);
SGVector<float64_t> lab_train(n);
feat_train(0,0)=-0.81263;
feat_train(0,1)=-0.99976;
feat_train(0,2)=1.17037;
feat_train(0,3)=-1.51752;
feat_train(0,4)=8.57765;
feat_train(0,5)=3.89440;
feat_train(1,0)=-0.5;
feat_train(1,1)=5.4576;
feat_train(1,2)=7.17637;
feat_train(1,3)=-2.56752;
feat_train(1,4)=4.57765;
feat_train(1,5)=2.89440;
lat_feat_train(0,0)=1.00000;
lat_feat_train(0,1)=23.00000;
lat_feat_train(0,2)=4.00000;
lat_feat_train(1,0)=3.00000;
lat_feat_train(1,1)=2.00000;
lat_feat_train(1,2)=-5.00000;
lab_train[0]=1;
lab_train[1]=-1;
lab_train[2]=1;
lab_train[3]=1;
lab_train[4]=-1;
lab_train[5]=-1;
// shogun representation of features and labels
CDenseFeatures<float64_t>* features_train=new CDenseFeatures<float64_t>(feat_train);
CDenseFeatures<float64_t>* latent_features_train=new CDenseFeatures<float64_t>(lat_feat_train);
CBinaryLabels* labels_train=new CBinaryLabels(lab_train);
float64_t ell=1.0;
CLinearARDKernel* kernel=new CGaussianARDFITCKernel(10, 2*ell*ell);
int32_t t_dim=2;
SGMatrix<float64_t> weights(t_dim,dim);
//the weights is a upper triangular matrix since GPML 3.5 only supports this type
float64_t weight1=0.02;
float64_t weight2=-0.4;
float64_t weight3=0;
float64_t weight4=0.01;
weights(0,0)=weight1;
weights(0,1)=weight2;
weights(1,0)=weight3;
weights(1,1)=weight4;
kernel->set_matrix_weights(weights);
float64_t mean_weight=2.0;
CConstMean* mean=new CConstMean(mean_weight);
CLogitLikelihood* lik=new CLogitLikelihood();
// specify GP with FITC inference
CSingleFITCLaplacianInferenceMethodWithLBFGS* inf=new CSingleFITCLaplacianInferenceMethodWithLBFGS(kernel, features_train,
mean, labels_train, lik, latent_features_train);
float64_t ind_noise=1e-6;
inf->set_inducing_noise(ind_noise);
float64_t scale=4.0;
inf->set_scale(scale);
// comparison of posterior cholesky with result from GPML 3.5 package:
// L =
//-0.079160062284788 0.043573099312206 0.003733704862708
//0.043573099312206 -0.062046601915619 -0.006797768274130
//0.003733704862708 -0.006797768274130 -0.006876868793106
SGMatrix<float64_t> L=inf->get_cholesky();
abs_tolorance = CMath::get_abs_tolerance(-0.079160062284788, rel_tolorance);
EXPECT_NEAR(L(0,0), -0.079160062284788, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.043573099312206, rel_tolorance);
EXPECT_NEAR(L(0,1), 0.043573099312206, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.003733704862708, rel_tolorance);
EXPECT_NEAR(L(0,2), 0.003733704862708, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.043573099312206, rel_tolorance);
EXPECT_NEAR(L(1,0), 0.043573099312206, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.062046601915619, rel_tolorance);
EXPECT_NEAR(L(1,1), -0.062046601915619, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.006797768274130, rel_tolorance);
EXPECT_NEAR(L(1,2), -0.006797768274130, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.003733704862708, rel_tolorance);
EXPECT_NEAR(L(2,0), 0.003733704862708, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.006797768274130, rel_tolorance);
EXPECT_NEAR(L(2,1), -0.006797768274130, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.006876868793106, rel_tolorance);
EXPECT_NEAR(L(2,2), -0.006876868793106, abs_tolorance);
SG_UNREF(inf);
}
TEST(SingleFITCLaplacianInferenceMethodWithLBFGS,get_alpha)
{
index_t n=6;
index_t dim=2;
index_t m=3;
float64_t rel_tolorance=1e-5;
float64_t abs_tolorance;
SGMatrix<float64_t> feat_train(dim, n);
SGMatrix<float64_t> lat_feat_train(dim, m);
SGVector<float64_t> lab_train(n);
feat_train(0,0)=-0.81263;
feat_train(0,1)=-0.99976;
feat_train(0,2)=1.17037;
feat_train(0,3)=-1.51752;
feat_train(0,4)=8.57765;
feat_train(0,5)=3.89440;
feat_train(1,0)=-0.5;
feat_train(1,1)=5.4576;
feat_train(1,2)=7.17637;
feat_train(1,3)=-2.56752;
feat_train(1,4)=4.57765;
feat_train(1,5)=2.89440;
lat_feat_train(0,0)=1.00000;
lat_feat_train(0,1)=23.00000;
lat_feat_train(0,2)=4.00000;
lat_feat_train(1,0)=3.00000;
lat_feat_train(1,1)=2.00000;
lat_feat_train(1,2)=-5.00000;
lab_train[0]=1;
lab_train[1]=-1;
lab_train[2]=1;
lab_train[3]=1;
lab_train[4]=-1;
lab_train[5]=-1;
// shogun representation of features and labels
CDenseFeatures<float64_t>* features_train=new CDenseFeatures<float64_t>(feat_train);
CDenseFeatures<float64_t>* latent_features_train=new CDenseFeatures<float64_t>(lat_feat_train);
CBinaryLabels* labels_train=new CBinaryLabels(lab_train);
float64_t ell=1.0;
CLinearARDKernel* kernel=new CGaussianARDFITCKernel(10, 2*ell*ell);
int32_t t_dim=2;
SGMatrix<float64_t> weights(t_dim,dim);
//the weights is a upper triangular matrix since GPML 3.5 only supports this type
float64_t weight1=0.02;
float64_t weight2=-0.4;
float64_t weight3=0;
float64_t weight4=0.01;
weights(0,0)=weight1;
weights(0,1)=weight2;
weights(1,0)=weight3;
weights(1,1)=weight4;
kernel->set_matrix_weights(weights);
float64_t mean_weight=2.0;
CConstMean* mean=new CConstMean(mean_weight);
CLogitLikelihood* lik=new CLogitLikelihood();
// specify GP with FITC inference
CSingleFITCLaplacianInferenceMethodWithLBFGS* inf=new CSingleFITCLaplacianInferenceMethodWithLBFGS(kernel, features_train,
mean, labels_train, lik, latent_features_train);
float64_t ind_noise=1e-6;
inf->set_inducing_noise(ind_noise);
float64_t scale=4.0;
inf->set_scale(scale);
// comparison of posterior alpha with result from GPML package:
// alpha =
//-0.366893026022809
//0.174674800097295
//0.028021450045659
SGVector<float64_t> alpha=inf->get_alpha();
abs_tolorance = CMath::get_abs_tolerance(-0.366893026022809, rel_tolorance);
EXPECT_NEAR(alpha[0], -0.366893026022809, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.174674800097295, rel_tolorance);
EXPECT_NEAR(alpha[1], 0.174674800097295, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.028021450045659, rel_tolorance);
EXPECT_NEAR(alpha[2], 0.028021450045659, abs_tolorance);
SG_UNREF(inf);
}
TEST(SingleFITCLaplacianInferenceMethodWithLBFGS,get_negative_log_marginal_likelihood)
{
index_t n=6;
index_t dim=2;
index_t m=3;
float64_t rel_tolorance=1e-5;
float64_t abs_tolorance;
SGMatrix<float64_t> feat_train(dim, n);
SGMatrix<float64_t> lat_feat_train(dim, m);
SGVector<float64_t> lab_train(n);
feat_train(0,0)=0.81263;
feat_train(0,1)=0.99976;
feat_train(0,2)=1.17037;
feat_train(0,3)=1.51752;
feat_train(0,4)=1.57765;
feat_train(0,5)=3.89440;
feat_train(1,0)=0.5;
feat_train(1,1)=0.4576;
feat_train(1,2)=5.17637;
feat_train(1,3)=2.56752;
feat_train(1,4)=4.57765;
feat_train(1,5)=2.89440;
lat_feat_train(0,0)=1;
lat_feat_train(0,1)=3;
lat_feat_train(0,2)=4;
lat_feat_train(1,0)=3;
lat_feat_train(1,1)=2;
lat_feat_train(1,2)=5;
lab_train[0]=1;
lab_train[1]=-1;
lab_train[2]=1;
lab_train[3]=1;
lab_train[4]=-1;
lab_train[5]=-1;
// shogun representation of features and labels
CDenseFeatures<float64_t>* features_train=new CDenseFeatures<float64_t>(feat_train);
CDenseFeatures<float64_t>* latent_features_train=new CDenseFeatures<float64_t>(lat_feat_train);
CBinaryLabels* labels_train=new CBinaryLabels(lab_train);
float64_t ell=1.0;
float64_t weight1=2.0;
float64_t weight2=3.0;
CLinearARDKernel* kernel=new CGaussianARDFITCKernel(10, 2*ell*ell);
SGVector<float64_t> weights(2);
weights[0]=1.0/weight1;
weights[1]=1.0/weight2;
kernel->set_vector_weights(weights);
CZeroMean* mean=new CZeroMean();
CLogitLikelihood* lik=new CLogitLikelihood();
// specify GP regression with FITC inference
CSingleFITCLaplacianInferenceMethodWithLBFGS* inf=new CSingleFITCLaplacianInferenceMethodWithLBFGS(kernel, features_train,
mean, labels_train, lik, latent_features_train);
float64_t ind_noise=1e-6;
inf->set_inducing_noise(ind_noise);
float64_t scale=5.0;
inf->set_scale(scale);
// comparison of posterior negative marginal likelihood with
// result from GPML 3.4 package using Newton method:
// nlZ =
// 5.479598775896901
float64_t nml=inf->get_negative_log_marginal_likelihood();
abs_tolorance = CMath::get_abs_tolerance(5.479598775896901, rel_tolorance);
EXPECT_NEAR(nml, 5.479598775896901, abs_tolorance);
SG_UNREF(inf);
}
TEST(SingleFITCLaplacianInferenceMethodWithLBFGS,get_marginal_likelihood_derivatives)
{
index_t n=6;
index_t dim=2;
index_t m=3;
float64_t rel_tolorance=1e-5;
float64_t abs_tolorance;
SGMatrix<float64_t> feat_train(dim, n);
SGMatrix<float64_t> lat_feat_train(dim, m);
SGVector<float64_t> lab_train(n);
feat_train(0,0)=0.81263;
feat_train(0,1)=0.99976;
feat_train(0,2)=1.17037;
feat_train(0,3)=1.51752;
feat_train(0,4)=1.57765;
feat_train(0,5)=3.89440;
feat_train(1,0)=0.5;
feat_train(1,1)=0.4576;
feat_train(1,2)=5.17637;
feat_train(1,3)=2.56752;
feat_train(1,4)=4.57765;
feat_train(1,5)=2.89440;
lat_feat_train(0,0)=1;
lat_feat_train(0,1)=3;
lat_feat_train(0,2)=4;
lat_feat_train(1,0)=3;
lat_feat_train(1,1)=2;
lat_feat_train(1,2)=5;
lab_train[0]=1;
lab_train[1]=-1;
lab_train[2]=1;
lab_train[3]=1;
lab_train[4]=-1;
lab_train[5]=-1;
// shogun representation of features and labels
CDenseFeatures<float64_t>* features_train=new CDenseFeatures<float64_t>(feat_train);
CDenseFeatures<float64_t>* latent_features_train=new CDenseFeatures<float64_t>(lat_feat_train);
CBinaryLabels* labels_train=new CBinaryLabels(lab_train);
float64_t ell=1.0;
float64_t weight1=2.0;
float64_t weight2=3.0;
CLinearARDKernel* kernel=new CGaussianARDFITCKernel(10, 2*ell*ell);
SGVector<float64_t> weights(2);
weights[0]=1.0/weight1;
weights[1]=1.0/weight2;
kernel->set_vector_weights(weights);
CZeroMean* mean=new CZeroMean();
CLogitLikelihood* lik=new CLogitLikelihood();
// specify GP regression with FITC inference
CSingleFITCLaplacianInferenceMethodWithLBFGS* inf=new CSingleFITCLaplacianInferenceMethodWithLBFGS(kernel, features_train,
mean, labels_train, lik, latent_features_train);
float64_t ind_noise=1e-6;
inf->set_inducing_noise(ind_noise);
float64_t scale=5.0;
inf->set_scale(scale);
// build parameter dictionary
CMap<TParameter*, CSGObject*>* parameter_dictionary=new CMap<TParameter*, CSGObject*>();
inf->build_gradient_parameter_dictionary(parameter_dictionary);
// compute derivatives wrt parameters
CMap<TParameter*, SGVector<float64_t> >* gradient=
inf->get_negative_log_marginal_likelihood_derivatives(parameter_dictionary);
// result from GPML 3.4 package:
//dnlz =
//mean: []
//cov: [0.889369895665482 0.260954643384094 0.579311763908738]
//lik: []
//dnlz.xu'
//0.140255088672368 -0.060040496377590 -0.157432965446797
//0.125896281241220 -0.067325533249551 -0.007553979244171
//
// get parameters to compute derivatives
TParameter* scale_param=inf->m_gradient_parameters->get_parameter("scale");
TParameter* weights_param=kernel->m_gradient_parameters->get_parameter("weights");
float64_t dnlz_weight1=(-1.0/weight1)*(gradient->get_element(weights_param))[0];
float64_t dnlz_weight2=(-1.0/weight2)*(gradient->get_element(weights_param))[1];
float64_t dnlZ_sf2=scale*(gradient->get_element(scale_param))[0];
TParameter* lat_param=inf->m_gradient_parameters->get_parameter("inducing_features");
SGVector<float64_t> dnlZ_lat=gradient->get_element(lat_param);
SGMatrix<float64_t> deriv_lat(dnlZ_lat.vector, dim, m, false);
abs_tolorance = CMath::get_abs_tolerance(0.579311763908738, rel_tolorance);
EXPECT_NEAR(dnlZ_sf2, 0.579311763908738, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.889369895665482, rel_tolorance);
EXPECT_NEAR(dnlz_weight1, 0.889369895665482, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.260954643384094, rel_tolorance);
EXPECT_NEAR(dnlz_weight2, 0.260954643384094, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.140255088672368, rel_tolorance);
EXPECT_NEAR(deriv_lat(0,0), 0.140255088672368, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.060040496377590, rel_tolorance);
EXPECT_NEAR(deriv_lat(0,1), -0.060040496377590, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.157432965446797, rel_tolorance);
EXPECT_NEAR(deriv_lat(0,2), -0.157432965446797, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(0.125896281241220, rel_tolorance);
EXPECT_NEAR(deriv_lat(1,0), 0.125896281241220, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.067325533249551, rel_tolorance);
EXPECT_NEAR(deriv_lat(1,1), -0.067325533249551, abs_tolorance);
abs_tolorance = CMath::get_abs_tolerance(-0.007553979244171, rel_tolorance);
EXPECT_NEAR(deriv_lat(1,2), -0.007553979244171, abs_tolorance);
// clean up
SG_UNREF(gradient);
SG_UNREF(parameter_dictionary);
SG_UNREF(inf);
}
#endif /* HAVE_EIGEN3 && HAVE_LINALG_LIB */