-
Notifications
You must be signed in to change notification settings - Fork 150
/
en2.c
869 lines (772 loc) · 28.2 KB
/
en2.c
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
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
/*
This program was automatically generated using:
__ ____ ____
/ / / / \ / __/ HBC: The Hierarchical Bayes Compiler
/ /_/ / / // / http://hal3.name/HBC/
/ __ / --</ /
/ / / / / / /___ Version 0.7 beta
\/ /_/____/\____/
HBC is a freely available compiler for statistical models. This generated
code can be built using the following command:
gcc -O3 -lm stats.c samplib.c en2.c -o en2.out
The hierarchical model that this code reflects is:
alphaC ~ Gam(1,1)
alphaD ~ Gam(1,1)
alphaE ~ Gam(0.1,1)
alphaV ~ Gam(0.1,1)
alphaW ~ Gam(0.1,1)
thetaC ~ DirSym(alphaC, Nc)
thetaD ~ DirSym(alphaD, Nd)
thetaE_{k,l} ~ DirSym(alphaE, Nen) , k \in [1,Nc] , l \in [1,Nd]
thetaV_{k} ~ DirSym(alphaV, VV) , k \in [1,Nc]
thetaW_{k} ~ DirSym(alphaW, VW) , k \in [1,Nen]
c_{n} ~ Mult(thetaC) , n \in [1,N]
d_{n} ~ Mult(thetaD) , n \in [1,N]
e_{n} ~ Mult(thetaE_{c_{n},d_{n}}) , n \in [1,N]
v_{n} ~ Mult(thetaV_{c_{n}}) , n \in [1,N]
w_{n} ~ Mult(thetaW_{e_{n}}) , n \in [1,N]
--# --define Nc 10
--# --define Nen 6
--# --define alphaC 2
--# --define alphaD 2
--# --define alphaE 0.001
--# --define alphaV 0.001
--# --define alphaW 0.001
--# --loadD enO w VW N ;
--# --loadD enV v VV N ;
--# --loadD enD d Nd N ;
--# --collapse thetaC
--# --collapse thetaD
--# --collapse thetaV
--# --collapse thetaW
Generated using the command:
hbc compile en2.hier en2.c
*/
#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include "stats.h"
/**************************** SAMPLING ****************************/
void resample_post_thetaC(int N, int Nc, int* c, double* post_thetaC) {
double* tmpSP9;
int n_5;
int dvv_loop_var_1;
tmpSP9 = (double*) malloc(sizeof(double) * (1+((Nc) + (1))-(1)));
/* Implements direct sampling from the following distribution: */
/* Delta(post_thetaC | \sum_{n@5 \in [N]} IDR(c_{n@5}, 1, Nc), Nc) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
tmpSP9[dvv_loop_var_1-1] = 0.0;
}
tmpSP9[(Nc) + (1)-1] = (0.0) * (((1) + (Nc)) - (1));
for (n_5=1; n_5<=N; n_5++) {
tmpSP9[(Nc) + (1)-1] += 1.0;
tmpSP9[c[n_5-1]-1] += 1.0;
}
sample_Delta(post_thetaC, tmpSP9, Nc);
free(tmpSP9);
}
void resample_post_thetaD(int N, int Nd, int* d, double* post_thetaD) {
double* tmpSP10;
int n_6;
int dvv_loop_var_1;
tmpSP10 = (double*) malloc(sizeof(double) * (1+((Nd) + (1))-(1)));
/* Implements direct sampling from the following distribution: */
/* Delta(post_thetaD | \sum_{n@6 \in [N]} IDR(d_{n@6}, 1, Nd), Nd) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nd; dvv_loop_var_1++) {
tmpSP10[dvv_loop_var_1-1] = 0.0;
}
tmpSP10[(Nd) + (1)-1] = (0.0) * (((1) + (Nd)) - (1));
for (n_6=1; n_6<=N; n_6++) {
tmpSP10[(Nd) + (1)-1] += 1.0;
tmpSP10[d[n_6-1]-1] += 1.0;
}
sample_Delta(post_thetaD, tmpSP10, Nd);
free(tmpSP10);
}
void resample_post_thetaV(int N, int Nc, int VV, int* c, double** post_thetaV, int* v) {
int k_23;
double* tmpSP12;
int n_8;
int dvv_loop_var_1;
tmpSP12 = (double*) malloc(sizeof(double) * (1+((VV) + (1))-(1)));
for (k_23=1; k_23<=Nc; k_23++) {
/* Implements direct sampling from the following distribution: */
/* Delta(post_thetaV_{k@23} | \sum_{n@8 \in [N]} .*(=(k@23, c_{n@8}), IDR(v_{n@8}, 1, VV)), VV) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=VV; dvv_loop_var_1++) {
tmpSP12[dvv_loop_var_1-1] = 0.0;
}
tmpSP12[(VV) + (1)-1] = (0.0) * (((1) + (VV)) - (1));
for (n_8=1; n_8<=N; n_8++) {
tmpSP12[(VV) + (1)-1] += (1.0) * ((((k_23) == (c[n_8-1])) ? 1 : 0));
tmpSP12[v[n_8-1]-1] += (1.0) * ((((k_23) == (c[n_8-1])) ? 1 : 0));
}
sample_Delta(post_thetaV[k_23-1], tmpSP12, VV);
}
free(tmpSP12);
}
void resample_post_thetaW(int N, int Nen, int VW, int* e, double** post_thetaW, int* w) {
int k_24;
double* tmpSP13;
int n_9;
int dvv_loop_var_1;
tmpSP13 = (double*) malloc(sizeof(double) * (1+((VW) + (1))-(1)));
for (k_24=1; k_24<=Nen; k_24++) {
/* Implements direct sampling from the following distribution: */
/* Delta(post_thetaW_{k@24} | \sum_{n@9 \in [N]} .*(=(k@24, e_{n@9}), IDR(w_{n@9}, 1, VW)), VW) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=VW; dvv_loop_var_1++) {
tmpSP13[dvv_loop_var_1-1] = 0.0;
}
tmpSP13[(VW) + (1)-1] = (0.0) * (((1) + (VW)) - (1));
for (n_9=1; n_9<=N; n_9++) {
tmpSP13[(VW) + (1)-1] += (1.0) * ((((k_24) == (e[n_9-1])) ? 1 : 0));
tmpSP13[w[n_9-1]-1] += (1.0) * ((((k_24) == (e[n_9-1])) ? 1 : 0));
}
sample_Delta(post_thetaW[k_24-1], tmpSP13, VW);
}
free(tmpSP13);
}
double resample_alphaC(int Nc, double alphaC, double* post_thetaC) {
double tmpSP0;
int cgds;
/* Implements direct sampling from the following distribution: */
/* Gam(alphaC | 1, /(1.0, -(1.0, /(1.0, \sum_{cgds \in [Nc]} log(.*(/(1.0, sub(.+(alphaC, post_thetaC), +(Nc, 1))), .+(alphaC, post_thetaC_{cgds}))))))) */
tmpSP0 = 0.0;
for (cgds=1; cgds<=Nc; cgds++) {
tmpSP0 += log(((1.0) / ((alphaC) + (post_thetaC[(Nc) + (1)-1]))) * ((alphaC) + (post_thetaC[cgds-1])));
}
alphaC = sample_Gam(1, (1.0) / ((1.0) - ((1.0) / (tmpSP0))));
return (alphaC);
}
double resample_alphaD(int Nd, double alphaD, double* post_thetaD) {
double tmpSP1;
int cgds;
/* Implements direct sampling from the following distribution: */
/* Gam(alphaD | 1, /(1.0, -(1.0, /(1.0, \sum_{cgds \in [Nd]} log(.*(/(1.0, sub(.+(alphaD, post_thetaD), +(Nd, 1))), .+(alphaD, post_thetaD_{cgds}))))))) */
tmpSP1 = 0.0;
for (cgds=1; cgds<=Nd; cgds++) {
tmpSP1 += log(((1.0) / ((alphaD) + (post_thetaD[(Nd) + (1)-1]))) * ((alphaD) + (post_thetaD[cgds-1])));
}
alphaD = sample_Gam(1, (1.0) / ((1.0) - ((1.0) / (tmpSP1))));
return (alphaD);
}
double resample_alphaE(int Nc, int Nd, int Nen, double alphaE, double*** thetaE) {
double tmpSP2;
int k_2;
int l_92;
int cgds;
/* Implements direct sampling from the following distribution: */
/* Gam(alphaE | 0.1, /(1.0, -(1.0, /(1.0, \sum_{k@2 \in [Nc]} \sum_{l@92 \in [Nd]} \sum_{cgds \in [Nen]} log(.*(/(1.0, sub(thetaE_{k@2,l@92}, +(Nen, 1))), thetaE_{k@2,l@92,cgds})))))) */
tmpSP2 = 0.0;
for (k_2=1; k_2<=Nc; k_2++) {
for (l_92=1; l_92<=Nd; l_92++) {
for (cgds=1; cgds<=Nen; cgds++) {
tmpSP2 += log(((1.0) / (thetaE[k_2-1][l_92-1][(Nen) + (1)-1])) * (thetaE[k_2-1][l_92-1][cgds-1]));
}
}
}
alphaE = sample_Gam(0.1, (1.0) / ((1.0) - ((1.0) / (tmpSP2))));
return (alphaE);
}
double resample_alphaV(int Nc, int VV, double alphaV, double** post_thetaV) {
double tmpSP5;
int k_3;
int cgds;
/* Implements direct sampling from the following distribution: */
/* Gam(alphaV | 0.1, /(1.0, -(1.0, /(1.0, \sum_{k@3 \in [Nc]} \sum_{cgds \in [VV]} log(.*(/(1.0, sub(.+(alphaV, post_thetaV_{k@3}), +(VV, 1))), .+(alphaV, post_thetaV_{k@3,cgds}))))))) */
tmpSP5 = 0.0;
for (k_3=1; k_3<=Nc; k_3++) {
for (cgds=1; cgds<=VV; cgds++) {
tmpSP5 += log(((1.0) / ((alphaV) + (post_thetaV[k_3-1][(VV) + (1)-1]))) * ((alphaV) + (post_thetaV[k_3-1][cgds-1])));
}
}
alphaV = sample_Gam(0.1, (1.0) / ((1.0) - ((1.0) / (tmpSP5))));
return (alphaV);
}
double resample_alphaW(int Nen, int VW, double alphaW, double** post_thetaW) {
double tmpSP7;
int k_4;
int cgds;
/* Implements direct sampling from the following distribution: */
/* Gam(alphaW | 0.1, /(1.0, -(1.0, /(1.0, \sum_{k@4 \in [Nen]} \sum_{cgds \in [VW]} log(.*(/(1.0, sub(.+(alphaW, post_thetaW_{k@4}), +(VW, 1))), .+(alphaW, post_thetaW_{k@4,cgds}))))))) */
tmpSP7 = 0.0;
for (k_4=1; k_4<=Nen; k_4++) {
for (cgds=1; cgds<=VW; cgds++) {
tmpSP7 += log(((1.0) / ((alphaW) + (post_thetaW[k_4-1][(VW) + (1)-1]))) * ((alphaW) + (post_thetaW[k_4-1][cgds-1])));
}
}
alphaW = sample_Gam(0.1, (1.0) / ((1.0) - ((1.0) / (tmpSP7))));
return (alphaW);
}
void resample_thetaE(int N, int Nc, int Nd, int Nen, double alphaE, int* c, int* d, int* e, double*** thetaE) {
int k_22;
int l_37;
double* tmpSP11;
int n_7;
double* vec_var_1;
double* vec_var_0;
int dvv_loop_var_1;
tmpSP11 = (double*) malloc(sizeof(double) * (1+((Nen) + (1))-(1)));
vec_var_1 = (double*) malloc(sizeof(double) * (1+((Nen) + (1))-(1)));
vec_var_0 = (double*) malloc(sizeof(double) * (1+((Nen) + (1))-(1)));
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nen; dvv_loop_var_1++) {
vec_var_0[dvv_loop_var_1-1] = alphaE;
}
vec_var_0[(Nen) + (1)-1] = (alphaE) * (((1) + (Nen)) - (1));
for (k_22=1; k_22<=Nc; k_22++) {
for (l_37=1; l_37<=Nd; l_37++) {
/* Implements direct sampling from the following distribution: */
/* Dir(thetaE_{k@22,l@37} | +(vec(alphaE, 1, Nen), \sum_{n@7 \in [N]} .*(=(l@37, d_{n@7}), .*(=(k@22, c_{n@7}), IDR(e_{n@7}, 1, Nen)))), Nen) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nen; dvv_loop_var_1++) {
tmpSP11[dvv_loop_var_1-1] = 0.0;
}
tmpSP11[(Nen) + (1)-1] = (0.0) * (((1) + (Nen)) - (1));
for (n_7=1; n_7<=N; n_7++) {
tmpSP11[(Nen) + (1)-1] += ((1.0) * ((((k_22) == (c[n_7-1])) ? 1 : 0))) * ((((l_37) == (d[n_7-1])) ? 1 : 0));
tmpSP11[e[n_7-1]-1] += ((1.0) * ((((k_22) == (c[n_7-1])) ? 1 : 0))) * ((((l_37) == (d[n_7-1])) ? 1 : 0));
}
sample_Dir(thetaE[k_22-1][l_37-1], add_vec_r_1(vec_var_1, vec_var_0, tmpSP11, 1, Nen), Nen);
}
}
free(tmpSP11);
free(vec_var_1);
free(vec_var_0);
}
void resample_c(int N, double alphaC, double alphaV, int* c, int* d, int* e, double* post_thetaC, double** post_thetaV, double*** thetaE, int* v, int Nc, int VV, int Nen) {
int n_25;
double* tmp_post_c_1;
int tmp_idx_c_1;
int dvv_loop_var_1;
tmp_post_c_1 = (double*) malloc(sizeof(double) * (1+((Nc) + (1))-(1)));
for (n_25=1; n_25<=N; n_25++) {
post_thetaV[c[n_25-1]-1][(VV) + (1)-1] += (0.0) - ((1.0) * ((((c[n_25-1]) == (c[n_25-1])) ? 1 : 0)));
post_thetaV[c[n_25-1]-1][v[n_25-1]-1] += (0.0) - ((1.0) * ((((c[n_25-1]) == (c[n_25-1])) ? 1 : 0)));
post_thetaC[(Nc) + (1)-1] += (0.0) - (1.0);
post_thetaC[c[n_25-1]-1] += (0.0) - (1.0);
/* Implements multinomial sampling from the following distribution: */
/* (Mult(e_{n@25} | sub(thetaE, c_{n@25}, d_{n@25})))((Mult(v_{n@25} | .+(alphaV, sub(post_thetaV, c_{n@25}))))(Mult(c_{n@25} | .+(alphaC, post_thetaC)))) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
tmp_post_c_1[dvv_loop_var_1-1] = 0.0;
}
tmp_post_c_1[(Nc) + (1)-1] = (0.0) * (((1) + (Nc)) - (1));
for (tmp_idx_c_1=1; tmp_idx_c_1<=Nc; tmp_idx_c_1++) {
tmp_post_c_1[tmp_idx_c_1-1] = (ldf_Mult(0, e[n_25-1], thetaE[tmp_idx_c_1-1][d[n_25-1]-1], 1, Nen)) + ((ldf_Mult_smooth(0, alphaV, v[n_25-1], post_thetaV[tmp_idx_c_1-1], 1, VV)) + (ldf_Mult_smooth(0, alphaC, tmp_idx_c_1, post_thetaC, 1, Nc)));
}
normalizeLog(tmp_post_c_1, 1, Nc);
c[n_25-1] = sample_Mult(tmp_post_c_1, 1, Nc);
post_thetaC[(Nc) + (1)-1] += 1.0;
post_thetaC[c[n_25-1]-1] += 1.0;
post_thetaV[c[n_25-1]-1][(VV) + (1)-1] += (1.0) * ((((c[n_25-1]) == (c[n_25-1])) ? 1 : 0));
post_thetaV[c[n_25-1]-1][v[n_25-1]-1] += (1.0) * ((((c[n_25-1]) == (c[n_25-1])) ? 1 : 0));
}
free(tmp_post_c_1);
}
void resample_d(int N, double alphaD, int* c, int* d, int* e, double* post_thetaD, double*** thetaE, int Nd, int Nen) {
int n_26;
double* tmp_post_d_1;
int tmp_idx_d_1;
int dvv_loop_var_1;
tmp_post_d_1 = (double*) malloc(sizeof(double) * (1+((Nd) + (1))-(1)));
for (n_26=1; n_26<=N; n_26++) {
post_thetaD[(Nd) + (1)-1] += (0.0) - (1.0);
post_thetaD[d[n_26-1]-1] += (0.0) - (1.0);
/* Implements multinomial sampling from the following distribution: */
/* (Mult(e_{n@26} | sub(thetaE, c_{n@26}, d_{n@26})))(Mult(d_{n@26} | .+(alphaD, post_thetaD))) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nd; dvv_loop_var_1++) {
tmp_post_d_1[dvv_loop_var_1-1] = 0.0;
}
tmp_post_d_1[(Nd) + (1)-1] = (0.0) * (((1) + (Nd)) - (1));
for (tmp_idx_d_1=1; tmp_idx_d_1<=Nd; tmp_idx_d_1++) {
tmp_post_d_1[tmp_idx_d_1-1] = (ldf_Mult(0, e[n_26-1], thetaE[c[n_26-1]-1][tmp_idx_d_1-1], 1, Nen)) + (ldf_Mult_smooth(0, alphaD, tmp_idx_d_1, post_thetaD, 1, Nd));
}
normalizeLog(tmp_post_d_1, 1, Nd);
d[n_26-1] = sample_Mult(tmp_post_d_1, 1, Nd);
post_thetaD[(Nd) + (1)-1] += 1.0;
post_thetaD[d[n_26-1]-1] += 1.0;
}
free(tmp_post_d_1);
}
void resample_e(int N, double alphaW, int* c, int* d, int* e, double** post_thetaW, double*** thetaE, int* w, int Nen, int VW) {
int n_27;
double* tmp_post_e_1;
int tmp_idx_e_1;
int dvv_loop_var_1;
tmp_post_e_1 = (double*) malloc(sizeof(double) * (1+((Nen) + (1))-(1)));
for (n_27=1; n_27<=N; n_27++) {
post_thetaW[e[n_27-1]-1][(VW) + (1)-1] += (0.0) - ((1.0) * ((((e[n_27-1]) == (e[n_27-1])) ? 1 : 0)));
post_thetaW[e[n_27-1]-1][w[n_27-1]-1] += (0.0) - ((1.0) * ((((e[n_27-1]) == (e[n_27-1])) ? 1 : 0)));
/* Implements multinomial sampling from the following distribution: */
/* (Mult(w_{n@27} | .+(alphaW, sub(post_thetaW, e_{n@27}))))(Mult(e_{n@27} | sub(thetaE, c_{n@27}, d_{n@27}))) */
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nen; dvv_loop_var_1++) {
tmp_post_e_1[dvv_loop_var_1-1] = 0.0;
}
tmp_post_e_1[(Nen) + (1)-1] = (0.0) * (((1) + (Nen)) - (1));
for (tmp_idx_e_1=1; tmp_idx_e_1<=Nen; tmp_idx_e_1++) {
tmp_post_e_1[tmp_idx_e_1-1] = (ldf_Mult_smooth(0, alphaW, w[n_27-1], post_thetaW[tmp_idx_e_1-1], 1, VW)) + (ldf_Mult(0, tmp_idx_e_1, thetaE[c[n_27-1]-1][d[n_27-1]-1], 1, Nen));
}
normalizeLog(tmp_post_e_1, 1, Nen);
e[n_27-1] = sample_Mult(tmp_post_e_1, 1, Nen);
post_thetaW[e[n_27-1]-1][(VW) + (1)-1] += (1.0) * ((((e[n_27-1]) == (e[n_27-1])) ? 1 : 0));
post_thetaW[e[n_27-1]-1][w[n_27-1]-1] += (1.0) * ((((e[n_27-1]) == (e[n_27-1])) ? 1 : 0));
}
free(tmp_post_e_1);
}
void resample_v(int N, double alphaV, int* c, double** post_thetaV, int* v, int VV) {
int n_28;
for (n_28=1; n_28<=N; n_28++) {
post_thetaV[c[n_28-1]-1][(VV) + (1)-1] += (0.0) - ((1.0) * ((((c[n_28-1]) == (c[n_28-1])) ? 1 : 0)));
post_thetaV[c[n_28-1]-1][v[n_28-1]-1] += (0.0) - ((1.0) * ((((c[n_28-1]) == (c[n_28-1])) ? 1 : 0)));
/* Implements direct sampling from the following distribution: */
/* Mult(v_{n@28} | .+(alphaV, sub(post_thetaV, c_{n@28}))) */
v[n_28-1] = sample_Mult_smooth(alphaV, post_thetaV[c[n_28-1]-1], 1, VV);
post_thetaV[c[n_28-1]-1][(VV) + (1)-1] += (1.0) * ((((c[n_28-1]) == (c[n_28-1])) ? 1 : 0));
post_thetaV[c[n_28-1]-1][v[n_28-1]-1] += (1.0) * ((((c[n_28-1]) == (c[n_28-1])) ? 1 : 0));
}
}
void resample_w(int N, double alphaW, int* e, double** post_thetaW, int* w, int VW) {
int n_29;
for (n_29=1; n_29<=N; n_29++) {
post_thetaW[e[n_29-1]-1][(VW) + (1)-1] += (0.0) - ((1.0) * ((((e[n_29-1]) == (e[n_29-1])) ? 1 : 0)));
post_thetaW[e[n_29-1]-1][w[n_29-1]-1] += (0.0) - ((1.0) * ((((e[n_29-1]) == (e[n_29-1])) ? 1 : 0)));
/* Implements direct sampling from the following distribution: */
/* Mult(w_{n@29} | .+(alphaW, sub(post_thetaW, e_{n@29}))) */
w[n_29-1] = sample_Mult_smooth(alphaW, post_thetaW[e[n_29-1]-1], 1, VW);
post_thetaW[e[n_29-1]-1][(VW) + (1)-1] += (1.0) * ((((e[n_29-1]) == (e[n_29-1])) ? 1 : 0));
post_thetaW[e[n_29-1]-1][w[n_29-1]-1] += (1.0) * ((((e[n_29-1]) == (e[n_29-1])) ? 1 : 0));
}
}
/************************* INITIALIZATION *************************/
double initialize_alphaC() {
double alphaC;
alphaC = sample_Gam(1.0, 1.0);
return (alphaC);
}
double initialize_alphaD() {
double alphaD;
alphaD = sample_Gam(1.0, 1.0);
return (alphaD);
}
double initialize_alphaE() {
double alphaE;
alphaE = sample_Gam(1.0, 1.0);
return (alphaE);
}
double initialize_alphaV() {
double alphaV;
alphaV = sample_Gam(1.0, 1.0);
return (alphaV);
}
double initialize_alphaW() {
double alphaW;
alphaW = sample_Gam(1.0, 1.0);
return (alphaW);
}
void initialize_thetaE(double*** thetaE, int Nc, int Nd, int Nen) {
int k_22;
int l_37;
int dvv_loop_var_1;
int dvv_loop_var_2;
int dvv_loop_var_3;
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=Nd; dvv_loop_var_2++) {
for (dvv_loop_var_3=1; dvv_loop_var_3<=Nen; dvv_loop_var_3++) {
thetaE[dvv_loop_var_1-1][dvv_loop_var_2-1][dvv_loop_var_3-1] = 0.0;
}
thetaE[dvv_loop_var_1-1][dvv_loop_var_2-1][(Nen) + (1)-1] = (0.0) * (((1) + (Nen)) - (1));
}
}
for (k_22=1; k_22<=Nc; k_22++) {
for (l_37=1; l_37<=Nd; l_37++) {
sample_DirSym(thetaE[k_22-1][l_37-1], 1.0, Nen);
}
}
}
void initialize_c(int* c, int N, int Nc) {
int n_25;
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
c[dvv_loop_var_1-1] = 0;
}
c[(N) + (1)-1] = (0) * (((1) + (N)) - (1));
for (n_25=1; n_25<=N; n_25++) {
c[n_25-1] = sample_MultSym(1, Nc);
}
}
void initialize_d(int* d, int N, int Nd) {
int n_26;
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
d[dvv_loop_var_1-1] = 0;
}
d[(N) + (1)-1] = (0) * (((1) + (N)) - (1));
for (n_26=1; n_26<=N; n_26++) {
d[n_26-1] = sample_MultSym(1, Nd);
}
}
void initialize_e(int* e, int N, int Nen) {
int n_27;
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
e[dvv_loop_var_1-1] = 0;
}
e[(N) + (1)-1] = (0) * (((1) + (N)) - (1));
for (n_27=1; n_27<=N; n_27++) {
e[n_27-1] = sample_MultSym(1, Nen);
}
}
void initialize_v(int* v, int N, int VV) {
int n_28;
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
v[dvv_loop_var_1-1] = 0;
}
v[(N) + (1)-1] = (0) * (((1) + (N)) - (1));
for (n_28=1; n_28<=N; n_28++) {
v[n_28-1] = sample_MultSym(1, VV);
}
}
void initialize_w(int* w, int N, int VW) {
int n_29;
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
w[dvv_loop_var_1-1] = 0;
}
w[(N) + (1)-1] = (0) * (((1) + (N)) - (1));
for (n_29=1; n_29<=N; n_29++) {
w[n_29-1] = sample_MultSym(1, VW);
}
}
void initialize_post_thetaC(double* post_thetaC, int N, int Nc, int* c) {
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
post_thetaC[dvv_loop_var_1-1] = 0.0;
}
post_thetaC[(Nc) + (1)-1] = (0.0) * (((1) + (Nc)) - (1));
resample_post_thetaC(N, Nc, c, post_thetaC);
}
void initialize_post_thetaD(double* post_thetaD, int N, int Nd, int* d) {
int dvv_loop_var_1;
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nd; dvv_loop_var_1++) {
post_thetaD[dvv_loop_var_1-1] = 0.0;
}
post_thetaD[(Nd) + (1)-1] = (0.0) * (((1) + (Nd)) - (1));
resample_post_thetaD(N, Nd, d, post_thetaD);
}
void initialize_post_thetaV(double** post_thetaV, int N, int Nc, int VV, int* c, int* v) {
int dvv_loop_var_1;
int dvv_loop_var_2;
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=VV; dvv_loop_var_2++) {
post_thetaV[dvv_loop_var_1-1][dvv_loop_var_2-1] = 0.0;
}
post_thetaV[dvv_loop_var_1-1][(VV) + (1)-1] = (0.0) * (((1) + (VV)) - (1));
}
resample_post_thetaV(N, Nc, VV, c, post_thetaV, v);
}
void initialize_post_thetaW(double** post_thetaW, int N, int Nen, int VW, int* e, int* w) {
int dvv_loop_var_1;
int dvv_loop_var_2;
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nen; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=VW; dvv_loop_var_2++) {
post_thetaW[dvv_loop_var_1-1][dvv_loop_var_2-1] = 0.0;
}
post_thetaW[dvv_loop_var_1-1][(VW) + (1)-1] = (0.0) * (((1) + (VW)) - (1));
}
resample_post_thetaW(N, Nen, VW, e, post_thetaW, w);
}
/**************************** DUMPING *****************************/
void dump_alphaC(double alphaC) {
printf("alphaC = ");
printf("%g", alphaC);
printf("\n");
}
void dump_alphaD(double alphaD) {
printf("alphaD = ");
printf("%g", alphaD);
printf("\n");
}
void dump_alphaE(double alphaE) {
printf("alphaE = ");
printf("%g", alphaE);
printf("\n");
}
void dump_alphaV(double alphaV) {
printf("alphaV = ");
printf("%g", alphaV);
printf("\n");
}
void dump_alphaW(double alphaW) {
printf("alphaW = ");
printf("%g", alphaW);
printf("\n");
}
void dump_thetaC(int Nc, double* thetaC) {
int dvv_loop_var_1;
printf("thetaC = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
printf("%g", thetaC[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_thetaD(int Nd, double* thetaD) {
int dvv_loop_var_1;
printf("thetaD = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nd; dvv_loop_var_1++) {
printf("%g", thetaD[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_thetaE(int Nc, int Nd, int Nen, double*** thetaE) {
int dvv_loop_var_1;
int dvv_loop_var_2;
int dvv_loop_var_3;
printf("thetaE = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=Nd; dvv_loop_var_2++) {
for (dvv_loop_var_3=1; dvv_loop_var_3<=Nen; dvv_loop_var_3++) {
printf("%g", thetaE[dvv_loop_var_1-1][dvv_loop_var_2-1][dvv_loop_var_3-1]);
printf(" ");
}
printf(" ; ");
}
printf(" ;; ");
}
printf("\n");
}
void dump_thetaV(int Nc, int VV, double** thetaV) {
int dvv_loop_var_1;
int dvv_loop_var_2;
printf("thetaV = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nc; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=VV; dvv_loop_var_2++) {
printf("%g", thetaV[dvv_loop_var_1-1][dvv_loop_var_2-1]);
printf(" ");
}
printf(" ; ");
}
printf("\n");
}
void dump_thetaW(int Nen, int VW, double** thetaW) {
int dvv_loop_var_1;
int dvv_loop_var_2;
printf("thetaW = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=Nen; dvv_loop_var_1++) {
for (dvv_loop_var_2=1; dvv_loop_var_2<=VW; dvv_loop_var_2++) {
printf("%g", thetaW[dvv_loop_var_1-1][dvv_loop_var_2-1]);
printf(" ");
}
printf(" ; ");
}
printf("\n");
}
void dump_c(int N, int* c) {
int dvv_loop_var_1;
printf("c = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
printf("%d", c[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_d(int N, int* d) {
int dvv_loop_var_1;
printf("d = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
printf("%d", d[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_e(int N, int* e) {
int dvv_loop_var_1;
printf("e = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
printf("%d", e[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_v(int N, int* v) {
int dvv_loop_var_1;
printf("v = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
printf("%d", v[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
void dump_w(int N, int* w) {
int dvv_loop_var_1;
printf("w = ");
for (dvv_loop_var_1=1; dvv_loop_var_1<=N; dvv_loop_var_1++) {
printf("%d", w[dvv_loop_var_1-1]);
printf(" ");
}
printf("\n");
}
/*************************** LIKELIHOOD ***************************/
double compute_log_posterior(int N, int Nc, int Nd, int Nen, int VV, int VW, double alphaC, double alphaD, double alphaE, double alphaV, double alphaW, int* c, int* d, int* e, double* thetaC, double* thetaD, double*** thetaE, double** thetaV, double** thetaW, int* v, int* w) {
double ldfP7_0;
double ldfP7_1;
int l_37;
int k_22;
double ldfP10_0;
int n_25;
double ldfP11_0;
int n_26;
double ldfP12_0;
int n_27;
double ldfP13_0;
int n_28;
double ldfP14_0;
int n_29;
ldfP7_0 = 0.0;
for (k_22=1; k_22<=Nc; k_22++) {
ldfP7_1 = 0.0;
for (l_37=1; l_37<=Nd; l_37++) {
ldfP7_1 += ldf_DirSym(1, thetaE[k_22-1][l_37-1], alphaE, Nen);
}
ldfP7_0 += ldfP7_1;
}
ldfP10_0 = 0.0;
for (n_25=1; n_25<=N; n_25++) {
ldfP10_0 += ldf_Mult(1, c[n_25-1], thetaC, 1, Nc);
}
ldfP11_0 = 0.0;
for (n_26=1; n_26<=N; n_26++) {
ldfP11_0 += ldf_Mult(1, d[n_26-1], thetaD, 1, Nd);
}
ldfP12_0 = 0.0;
for (n_27=1; n_27<=N; n_27++) {
ldfP12_0 += ldf_Mult(1, e[n_27-1], thetaE[c[n_27-1]-1][d[n_27-1]-1], 1, Nen);
}
ldfP13_0 = 0.0;
for (n_28=1; n_28<=N; n_28++) {
ldfP13_0 += ldf_Mult(1, v[n_28-1], thetaV[c[n_28-1]-1], 1, VV);
}
ldfP14_0 = 0.0;
for (n_29=1; n_29<=N; n_29++) {
ldfP14_0 += ldf_Mult(1, w[n_29-1], thetaW[e[n_29-1]-1], 1, VW);
}
return ((ldf_Gam(1, alphaC, 1, 1)) + ((ldf_Gam(1, alphaD, 1, 1)) + ((ldf_Gam(1, alphaE, 0.1, 1)) + ((ldf_Gam(1, alphaV, 0.1, 1)) + ((ldf_Gam(1, alphaW, 0.1, 1)) + ((0.0) + ((0.0) + ((ldfP7_0) + ((0.0) + ((0.0) + ((ldfP10_0) + ((ldfP11_0) + ((ldfP12_0) + ((ldfP13_0) + (ldfP14_0)))))))))))))));
}
/****************************** MAIN ******************************/
int main(int ARGC, char *ARGV[]) {
double loglik,bestloglik;
int iter;
int N;
int Nc;
int Nd;
int Nen;
int VV;
int VW;
double alphaC;
double alphaD;
double alphaE;
double alphaV;
double alphaW;
int* c;
int* d;
int* e;
double* post_thetaC;
double* post_thetaD;
double** post_thetaV;
double** post_thetaW;
double*** thetaE;
int* v;
int* w;
int malloc_dim_1;
int malloc_dim_2;
fprintf(stderr, "-- This program was automatically generated using HBC (v 0.7 beta) from en2.hier\n-- see http://hal3.name/HBC for more information\n");
fflush(stderr);
setall(time(0),time(0)); /* initialize random number generator */
/* variables defined with --define */
Nc = 100;
Nen = 40;
alphaC = 1;
alphaD = 1;
alphaE = 1.0e-5;
alphaV = 1.0e-3;
alphaW = 1.0e-3;
fprintf(stderr, "Loading data...\n");
fflush(stderr);
/* variables defined with --loadD */
w = load_discrete1("enO", &N, &VW);
v = load_discrete1("enV", &N, &VV);
d = load_discrete1("enD", &N, &Nd);
printf ("sizes %d %d %d %d\n",N,VW,VV,Nd);
printf ("debug V %d\n",v[154623]);
/* variables defined with --loadM or --loadMI */
fprintf(stderr, "Allocating memory...\n");
fflush(stderr);
c = (int*) malloc(sizeof(int) * (1+((N) + (1))-(1)));
e = (int*) malloc(sizeof(int) * (1+((N) + (1))-(1)));
post_thetaC = (double*) malloc(sizeof(double) * (1+((Nc) + (1))-(1)));
post_thetaD = (double*) malloc(sizeof(double) * (1+((Nd) + (1))-(1)));
post_thetaV = (double**) malloc(sizeof(double*) * (1+(Nc)-(1)));
for (malloc_dim_1=1; malloc_dim_1<=Nc; malloc_dim_1++) {
post_thetaV[malloc_dim_1-1] = (double*) malloc(sizeof(double) * (1+((VV) + (1))-(1)));
}
post_thetaW = (double**) malloc(sizeof(double*) * (1+(Nen)-(1)));
for (malloc_dim_1=1; malloc_dim_1<=Nen; malloc_dim_1++) {
post_thetaW[malloc_dim_1-1] = (double*) malloc(sizeof(double) * (1+((VW) + (1))-(1)));
}
thetaE = (double***) malloc(sizeof(double**) * (1+(Nc)-(1)));
for (malloc_dim_1=1; malloc_dim_1<=Nc; malloc_dim_1++) {
thetaE[malloc_dim_1-1] = (double**) malloc(sizeof(double*) * (1+(Nd)-(1)));
for (malloc_dim_2=1; malloc_dim_2<=Nd; malloc_dim_2++) {
thetaE[malloc_dim_1-1][malloc_dim_2-1] = (double*) malloc(sizeof(double) * (1+((Nen) + (1))-(1)));
}
}
fprintf(stderr, "Initializing variables...\n");
fflush(stderr);
initialize_thetaE(thetaE, Nc, Nd, Nen);
initialize_c(c, N, Nc);
initialize_e(e, N, Nen);
initialize_post_thetaC(post_thetaC, N, Nc, c);
initialize_post_thetaD(post_thetaD, N, Nd, d);
initialize_post_thetaV(post_thetaV, N, Nc, VV, c, v);
initialize_post_thetaW(post_thetaW, N, Nen, VW, e, w);
for (iter=1; iter<=100; iter++) {
fprintf(stderr, "iter %d", iter);
fflush(stderr);
resample_thetaE(N, Nc, Nd, Nen, alphaE, c, d, e, thetaE);
resample_c(N, alphaC, alphaV, c, d, e, post_thetaC, post_thetaV, thetaE, v, Nc, VV, Nen);
resample_e(N, alphaW, c, d, e, post_thetaW, thetaE, w, Nen, VW);
loglik = compute_log_posterior(N, Nc, Nd, Nen, VV, VW, alphaC, alphaD, alphaE, alphaV, alphaW, c, d, e, post_thetaC, post_thetaD, thetaE, post_thetaV, post_thetaW, v, w);
fprintf(stderr, "\t%g", loglik);
if ((iter==1)||(loglik>bestloglik)) {
bestloglik = loglik;
fprintf(stderr, " *");
}
printf("\n");
dump_c(N,c);
dump_e(N,e);
fprintf(stderr, "\n");
fflush(stderr);
}
free(w);
free(v);
for (malloc_dim_1=1; malloc_dim_1<=Nc; malloc_dim_1++) {
for (malloc_dim_2=1; malloc_dim_2<=Nd; malloc_dim_2++) {
free(thetaE[malloc_dim_1-1][malloc_dim_2-1]);
}
free(thetaE[malloc_dim_1-1]);
}
free(thetaE);
for (malloc_dim_1=1; malloc_dim_1<=Nen; malloc_dim_1++) {
free(post_thetaW[malloc_dim_1-1]);
}
free(post_thetaW);
for (malloc_dim_1=1; malloc_dim_1<=Nc; malloc_dim_1++) {
free(post_thetaV[malloc_dim_1-1]);
}
free(post_thetaV);
free(post_thetaD);
free(post_thetaC);
free(e);
free(d);
free(c);
return 0;
}