-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathglim_image.c
3328 lines (2818 loc) · 123 KB
/
glim_image.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
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/* ----------------------------- MNI Header -----------------------------------
@NAME : glim_image
@INPUT : argc, argv - command line arguments
@OUTPUT : (nothing)
@RETURNS : status
@DESCRIPTION: Program to fit a generalized linear statistical model to a
set of minc images and output various diagnostic images,
@CREATED : June 18, 1997 (J. Taylor)
@MODIFIED :
---------------------------------------------------------------------------- */
#if HAVE_CONFIG_H
#include "config.h"
#endif
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <float.h>
#include <math.h>
#include <voxel_loop.h>
#include <time_stamp.h>
#include <ParseArgv.h>
#include <volume_io.h>
#include "glim.h"
#include "smoothness.h"
#define BOOLEAN_DEFAULT (-1)
#define INVALID_DATA (-DBL_MAX)
/* typedefs */
typedef struct {
Glm_Object *glm_obj;
Lambda *lambda_buffer;
char **tmp_files;
} Program_Data;
/* Function prototypes */
int get_t_stat_raw(char *dst, char *key, int argc,
char **argv);
int get_f_stat_raw(char *dst, char *key, int argc,
char **argv);
int get_avg_raw(char *dst, char *key, int argc,
char **argv);
void voxel_function_glm(void *caller_data, long num_voxels,
int input_num_buffers, int input_vector_length,
double *input_data[],
int output_num_buffers, int output_vector_length,
double *output_data[],
Loop_Info *loop_info);
void voxel_function_pool_con(void *caller_data, long num_voxels,
int input_num_buffers, int input_vector_length,
double *input_data[],
int output_num_buffers, int output_vector_length,
double *output_data[],
Loop_Info *loop_info);
void voxel_function_pool_avg(void *caller_data, long num_voxels,
int input_num_buffers, int input_vector_length,
double *input_data[],
int output_num_buffers, int output_vector_length,
double *output_data[],
Loop_Info *loop_info);
void start_function(void *caller_data, long num_voxels,
int output_num_buffers, int output_vector_length,
double *output_data[], Loop_Info *loop_info);
void end_function(void *caller_data, long num_voxels,
int output_num_buffers, int output_vector_length,
double *output_data[], Loop_Info *loop_info);
void create_stat_variable(Program_Data *program_data, char **input_files);
void create_glim_parameters(Glm_Object *glm_obj, char **input_files,
char **param_str);
void get_fwhm_filenames(Lambda *lambda_buffer, Glm_Object *glm_obj);
int update_lambda_buffer(Lambda *lambda_buffer, Glm_Object *glm_obj,
double check);
/* Argument variables */
/* Glim_image specific variables */
char *family_str = NULL;
char *link_str = NULL;
char *variance_function_str = NULL;
char *corr_struct_str = NULL;
int use_mu_switch = FALSE;
int binomial_n = 1;
int num_response = 1;
int max_num_row = 300;
int max_num_col = 100;
static Contrast_Raw_Array contrast_raw_list = {0, NULL};
double tol = 0.0001;
double eps_sing = 0.0001;
double max_iter= MAX_ITERATIONS;
double scale_par = -1;
double maxVoxelSdBasedValue = 99.0;
int est_scale = FALSE;
int chi_sq_nu = 0;
int is_fwhm_flag = FALSE;
int fit_all = FALSE;
int scylla = FALSE;
char *deg_free_file = NULL;
char *deviance_file = NULL;
char *avg_deviance_file = NULL;
char *scale_file = NULL;
char *pool_mask = NULL;
char *search_mask = NULL;
char *fwhm_gaussian_file = NULL;
char *fwhm_simple_file = NULL;
char *fwhm_avg_file = NULL;
char *error_file = NULL;
char *path_tmp = NULL;
FILE *tmpfile_list = NULL;
char *tmpfile_name = NULL;
int num_tmpfiles = 0;
static long num_fitted = 0;
static long num_skipped = 0;
static long num_singular = 0;
/* Voxel loop specific variables */
int clobber = FALSE;
int verbose = TRUE;
nc_type datatype = NC_UNSPECIFIED;
int is_signed = TRUE;
double valid_range[2] = {0,0};
int copy_all_header = FALSE;
int max_buffer_size_in_kb = 10 * 1024;
int check_dim_info = TRUE;
/* Argument table */
ArgvInfo argTable[] = {
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"Stats options:"},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"Common options:"},
{"-family", ARGV_STRING, (char *) NULL, (char *) &family_str,
"Specify exponential family for GLM."},
{"-link", ARGV_STRING, (char *) NULL, (char *) &link_str,
"Specify link function for GLM."},
{"-var_func", ARGV_STRING, (char *) NULL, (char *) &variance_function_str,
"Specify variance function for GLM."},
{"-t_stat", ARGV_GENFUNC, (char *) get_t_stat_raw,
(char *) &contrast_raw_list,
"Defines t-statistics wanted for output"},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"T-statistics are defined as follows: -t_stat <out.mnc> type (one of t_stat, \nstdev, corr, beta) stdev_type (one of pool or voxel) intype (one of matrix, \ncolumn or file) string (string defining contrast depends on intype)"},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"For example, the following: \n -t_stat out.mnc corr pool column 4 \ndefines an output file that will be a partial correlation testing column 4 \nof the design matrix"},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
" -t_stat out.mnc beta voxel matrix '0 1.5 0 -2.3' \ndefines an output file that will have the of 1.5*beta2 - 2.3*beta4 \n where the betas are the parameters estimated in the regression."},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
" -t_stat out.mnc stdev pool file contrast1.txt \ndefines an output file that will be the stdev of the t_stat defined in the \ntextfile contrast1.txt with a pooled stdev. This doesn't make sense because \nwe would not use a pooled stdev to look at the stdev as it varies over the\nimage, so glim_image spits this out as a warning."},
{"-f_stat", ARGV_GENFUNC, (char *) get_f_stat_raw,
(char *) &contrast_raw_list,
"Defines F-statistics wanted for output"},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"F_statistics are defined in the same way as t_statistics except there is no \ntype, this is because the variance of an F-stat is a matrix, as is the \nnumerator. For this reason, column is also not an intype for F_statistics \nbecause a matrix with more than two rows is needed for an F_stat."},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"For example: \n -f_stat out.mnc pool matrix '0 1 2 0 ; 1 2 3 0' \ndefines an output file that is an F-stat testing whether \n beta2 + 2*beta3 = 0 and beta1 + beta2 + beta3 = 0 \nsimultaneously. This comes into play for anova designs when there are factors\nwith multiple levels."},
{"-est_scale", ARGV_CONSTANT, (char *) TRUE, (char *) &est_scale,
"Use estimated scale (for binomial and poisson families)"},
{"-scale_par", ARGV_FLOAT, (char *) 1, (char *) &scale_par,
"Set scale parameter, (default: estimate it)"},
{"-max_iter", ARGV_FLOAT, (char *) 1, (char *) &max_iter,
"Maximum iterations for Fisher scoring."},
{"-tolerance", ARGV_FLOAT, (char *) 1, (char *) &tol,
"Tolerance for convergence of fitted parameters."},
{"-eps_sing", ARGV_FLOAT, (char *) 1, (char *) &eps_sing,
"Epsilon added to repsonse values that would cause \nsingularities."},
{"-maxVoxelSdBasedValue", ARGV_FLOAT, (char *) 1,
(char *) &maxVoxelSdBasedValue, "Maximum absolute value of t_statistic image."},
{"-avg", ARGV_GENFUNC, (char *) get_avg_raw, (char *) &contrast_raw_list,
"File name for means (missing values not skipped. Needs sd_type and out_type."},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"Avg is used for comparison of means at the same time as fitting a design\n file, the output types are t_stat, beta, corr and stdev; stdev types are \nvoxel or pool."},
{"-deviance", ARGV_STRING, (char *) NULL, (char *) &deviance_file,
"Output an image of deviances (default: FALSE)"},
{"-avg_deviance", ARGV_STRING, (char *) NULL, (char *) &avg_deviance_file,
"Output image of deviances for avg."},
{"-scale", ARGV_STRING, (char *) NULL, (char *) &scale_file,
"Output image of scale estimate (default: FALSE)"},
{"-errfile", ARGV_STRING, (char *) NULL, (char *) &error_file,
"Output convergence error codes."},
{"-use_mu", ARGV_CONSTANT, (char *) TRUE, (char *) &use_mu_switch,
"Use previous fitted values instead of response for \ninitial values."},
{"-pool_sd", ARGV_STRING, (char *) NULL, (char *) &pool_mask,
"Mask file for region over which to pool s.d."},
{"-fwhm", ARGV_CONSTANT, (char *) TRUE, (char *) &is_fwhm_flag,
"Calculate FWHM estimate to be used for assessing significance."},
{"-fwhm_gauss", ARGV_STRING, (char *) NULL, (char *) &fwhm_gaussian_file,
"Output fwhm image for gaussian fit."},
{"-fwhm_simple", ARGV_STRING, (char *) NULL, (char *) &fwhm_simple_file,
"Output non-weight corrected FWHM estimate for non-Gaussian data."},
{"-fwhm_avg", ARGV_STRING, (char *) NULL, (char *) &fwhm_avg_file,
"Output fwhm image for comparison of means fit."},
{"-search", ARGV_STRING, (char *) NULL, (char *) &search_mask,
"Mask file for search region."},
{"-fit_all", ARGV_CONSTANT, (char *) TRUE, (char *) &fit_all,
"Fit GLM at all voxels with no missing values, not just search region."},
{"-fit_missing", ARGV_STRING, (char *) TRUE, (char *) °_free_file,
"Fit GLM at all voxels, even with missing values. Must provide name for\ndegrees of freedom volume."},
{"-max_num_row", ARGV_INT, (char *) 1, (char *) &max_num_row,
"Maximum number of rows for design_file. Should be \nincreased for large files."},
{"-max_num_col", ARGV_INT, (char *) 1, (char *) &max_num_col,
"Maximum number of rows for design_file. Should be \nincreased for large files."},
{"-scylla", ARGV_CONSTANT, (char *) TRUE, (char *) &scylla,
"Flag to see if computer is at neuro or math dept."},
{NULL, ARGV_HELP, (char *) NULL, (char *) NULL,
"General options:"},
{"-path", ARGV_STRING, (char *) NULL, (char *) &path_tmp,
"Path for input files and design_matrix file."},
{"-clobber", ARGV_CONSTANT, (char *) TRUE, (char *) &clobber,
"Overwrite existing file."},
{"-noclobber", ARGV_CONSTANT, (char *) FALSE, (char *) &clobber,
"Don't overwrite existing file (default)."},
{"-verbose", ARGV_CONSTANT, (char *) TRUE, (char *) &verbose,
"Print out log messages (default)."},
{"-quiet", ARGV_CONSTANT, (char *) FALSE, (char *) &verbose,
"Do not print out log messages."},
{"-copy_header", ARGV_CONSTANT, (char *) TRUE, (char *) ©_all_header,
"Copy all of the header from the first file."},
{"-nocopy_header", ARGV_CONSTANT, (char *) FALSE, (char *) ©_all_header,
"Do not copy all of the header from the first file."},
{"-filetype", ARGV_CONSTANT, (char *) NC_UNSPECIFIED, (char *) &datatype,
"Use data type of first file (default)."},
{"-byte", ARGV_CONSTANT, (char *) NC_BYTE, (char *) &datatype,
"Write out byte data."},
{"-short", ARGV_CONSTANT, (char *) NC_SHORT, (char *) &datatype,
"Write out short integer data."},
{"-long", ARGV_CONSTANT, (char *) NC_LONG, (char *) &datatype,
"Write out long integer data."},
{"-float", ARGV_CONSTANT, (char *) NC_FLOAT, (char *) &datatype,
"Write out single-precision floating-point data."},
{"-double", ARGV_CONSTANT, (char *) NC_DOUBLE, (char *) &datatype,
"Write out double-precision floating-point data."},
{"-signed", ARGV_CONSTANT, (char *) TRUE, (char *) &is_signed,
"Write signed integer data."},
{"-unsigned", ARGV_CONSTANT, (char *) FALSE, (char *) &is_signed,
"Write unsigned integer data (default if type specified)."},
{"-range", ARGV_FLOAT, (char *) 2, (char *) valid_range,
"Valid range for output data."},
{"-max_buffer_size_in_kb", ARGV_INT, (char *) 1,
(char *) &max_buffer_size_in_kb,
"Specify the maximum size of the internal buffers (in kbytes)."},
{"-check_dimensions", ARGV_CONSTANT, (char *) TRUE,
(char *) &check_dim_info,
"Check that files have matching dimensions (default)."},
{"-nocheck_dimensions", ARGV_CONSTANT, (char *) FALSE,
(char *) &check_dim_info,
"Do not check that files have matching dimensions."},
{NULL, ARGV_END, NULL, NULL, NULL}
};
/* Main program */
int main(int argc, char *argv[])
{
char *prog, **input_files, **output_files, **infiles_pool;
char *design_filename;
char *path_in = NULL;
char *path_end = "/";
int num_input_files, num_output_files;
int path_len;
Loop_Options *loop_options;
Matrix *design_matrix;
Matrix *tmp, *tmp2, *singval, *buffer;
Contrast_Mat_Array *contrast_matrices;
char *arg_string, *tmp_count;
int i, j, pool_test, is_avg_fwhm;
Contrast_Matrix *cur_matrix, *cur_matrix2;
Avg_Info *cur_avg, *cur_avg2;
Program_Data *program_data;
Glm_Object *glm_obj;
Initial_Data *initial_data;
char **family_names, **link_names, **variance_function_names;
char **corr_struct_names;
Family family_input;
Link link_input;
Variance_Function variance_function_input;
Corr_Struct corr_input;
Lambda *lambda_buffer = NULL;
Volume tmp_volume;
volume_input_struct volume_input;
int *sizes;
int num_dim, itest, is_fwhm;
int volume_sizes[MAX_DIMENSIONS];
double volume_step[MAX_DIMENSIONS];
Fwhm_Info *fwhm_general, *fwhm_gaussian;
tmpfile_name = tempnam(NULL, "glim");
tmpfile_list = fopen(tmpfile_name, "w");
/* Save time stamp and args */
arg_string = time_stamp(argc, argv);
/* Check arguments */
prog = argv[0];
if (ParseArgv(&argc, argv, argTable, 0) || (argc != 2)) {
(void) fprintf(stderr, "\nUsage: %s [options] <des.file> \n", prog);
(void) fprintf(stderr,
" %s -help\n", prog);
fprintf(stderr,"\nYour command line was (after parsing): \n\n ");
for(i=0; i<argc; i++) {
fprintf(stderr,"%s ",argv[i]);
}
fprintf(stderr,"\n\n");
fprintf(stderr,"Use '-' as des.file to read from stdin. Last line should end with 'END'.\n\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
create_family_names(&family_names, &link_names, &variance_function_names,
&corr_struct_names);
family_input = get_family(family_str, family_names);
link_input = get_link(link_str, link_names);
variance_function_input = get_variance_function(variance_function_str,
variance_function_names);
corr_input = get_corr_struct(corr_struct_str, corr_struct_names);
/* Check to see that path ends in '/' */
if(path_tmp != NULL) {
path_len = strlen(path_tmp);
if(path_tmp[path_len-1] != path_end[0]) {
path_in = GI_MALLOC(sizeof(char) * (path_len + 2));
strcpy(path_in, path_tmp);
strcat(path_in, path_end);
}
}
/* Get design matrix from file design_filename */
design_filename = argv[1];
if( strcmp(design_filename,"-") == 0){
design_filename = NULL;
}
if(get_design_matrix(design_filename, &input_files, &design_matrix,
path_in, num_response, max_num_row,
max_num_col) != TRUE) {
fprintf(stderr,"Error: error in get_design_matrix_file. \n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
/* a bit of output */
if(verbose){
fprintf(stdout, "+++ design matrix from %s +++\n", design_filename);
for(i=0; i<design_matrix->num_rows; i++) {
fprintf(stdout, " | [%s] - ", input_files[i]);
for(j=0; j<design_matrix->num_columns; j++) {
fprintf(stdout, "%5g ", design_matrix->values[i][j]);
}
fprintf(stdout, "\n");
}
}
contrast_matrices = GI_MALLOC(sizeof(Contrast_Mat_Array));
contrast_matrices->num_columns = design_matrix->num_columns;
contrast_matrices->num_contrasts = (contrast_raw_list.num_output -
contrast_raw_list.num_avg);
contrast_matrices->num_avg = contrast_raw_list.num_avg;
contrast_matrices->contrast_matrix_array =\
GI_MALLOC(contrast_matrices->num_contrasts * sizeof(Matrix *));
contrast_matrices->avg_array = GI_MALLOC(sizeof(Avg_Array));
get_contrast_from_raw(&contrast_raw_list, &contrast_matrices);
/* Check ParseArgv options */
if ((fwhm_gaussian_file != NULL) || (fwhm_avg_file != NULL) ||
(fwhm_simple_file != NULL) || (is_fwhm_flag == TRUE))
is_fwhm = TRUE;
if ((fwhm_gaussian_file != NULL) && (fwhm_simple_file != NULL)) {
fprintf(stderr,"\nError: \"-fwhm_gauss\" and \"-fwhm_simple\" produce same output. \"-fwhm_gauss\" is only if data is Gaussian.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
if ((fwhm_gaussian_file != NULL) && (family_input != GAUSSIAN)) {
fprintf(stderr,"\nError: \"-fwhm_gauss\" option called for when family is not gaussian, use \"-fwhm_simple\".\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
if ((contrast_matrices->avg_array == NULL) && (fwhm_avg_file != NULL)){
fprintf(stderr,"\nError: \"-fwhm_avg\" option called for when no average test statistics being output.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
else if (contrast_matrices->avg_array != NULL) {
if ((contrast_matrices->avg_array->num_test <= 0)
&& (fwhm_avg_file != NULL)){
fprintf(stderr,"\nError: \"-fwhm_avg\" option called for when no average test statistics being output.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
if (family_input == GAUSSIAN) {
if (contrast_matrices->avg_array == NULL) {
if ((contrast_matrices->num_test <= 0) && (fwhm_gaussian_file != NULL)) {
fprintf(stderr,"\nError: \"-fwhm_gauss\" option called for when no test statistics being output\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
else if (contrast_matrices->avg_array->num_test <= 0) {
if ((contrast_matrices->num_test <= 0) && (fwhm_gaussian_file != NULL)) {
fprintf(stderr,"\nError: \"-fwhm_gauss\" option called for when no test statistics being output\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
else {
if (contrast_matrices->avg_array == NULL) {
if ((contrast_matrices->num_test <= 0) && (is_fwhm_flag == TRUE)) {
fprintf(stderr,"\nError: \"-fwhm\" option called for when no test statistics being output\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
else if (contrast_matrices->avg_array->num_test <= 0) {
if ((contrast_matrices->num_test <= 0) && (is_fwhm_flag == TRUE)) {
fprintf(stderr,"\nError: \"-fwhm\" option called for when no test statistics being output\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
/* Initialize lambda_buffer first because of its size */
if (is_fwhm == TRUE) {
if(start_volume_input(input_files[0], 3, NULL, NC_UNSPECIFIED, FALSE,
0.0, 0.0, TRUE, &tmp_volume,
(minc_input_options *) NULL, &volume_input) != OK){
fprintf(stderr,"\nError opening %s to check for size[] and step[] for smoothness calculation.\n", input_files[0]);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
num_dim = get_volume_n_dimensions(tmp_volume);
get_volume_sizes(tmp_volume, volume_sizes);
get_volume_separations(tmp_volume, volume_step);
cancel_volume_input(tmp_volume, &volume_input);
sizes = GI_MALLOC(sizeof(*sizes) * num_dim);
for(i=0; i<num_dim; i++) {
sizes[i] = volume_sizes[i];
}
if (contrast_matrices->avg_array != NULL) {
if(contrast_matrices->avg_array->num_test > 0)
is_avg_fwhm = TRUE;
else
is_avg_fwhm = FALSE;
}
else
is_avg_fwhm = FALSE;
lambda_buffer = create_lambda_buffer(design_matrix->num_rows,
contrast_matrices->num_test,
num_dim, sizes,
(family_input == GAUSSIAN),
is_avg_fwhm, fwhm_simple_file);
if(lambda_buffer == NULL) {
fprintf(stderr,"\nMemory error in create_lambda_buffer.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
lambda_buffer->constant = sqrt(4.0 * log(2.0));
for(i=0; i<num_dim; i++) {
lambda_buffer->constant *= pow(volume_step[i], 1.0/num_dim);
lambda_buffer->step[i] = volume_step[i];
}
}
initial_data = GI_MALLOC(sizeof(*initial_data));
initial_data->design_matrix = design_matrix;
initial_data->contrast = contrast_matrices;
initial_data->family = family_input;
initial_data->link = link_input;
initial_data->variance_function = variance_function_input;
initial_data->corr_struct = corr_input;
/* Get control parameters */
initial_data->control = GI_MALLOC(sizeof(Glm_Control));
if(pool_mask != NULL)
initial_data->control->pool_id = TRUE;
else
initial_data->control->pool_id = FALSE;
initial_data->control->status = CONT;
initial_data->control->scylla = scylla;
if(search_mask != NULL)
initial_data->control->search_id = TRUE;
else
initial_data->control->search_id = FALSE;
initial_data->control->fit_all = fit_all;
if (deg_free_file != NULL) {
initial_data->control->fit_all = TRUE;
initial_data->control->deg_free_id = TRUE;
}
else
initial_data->control->deg_free_id = -1;
if(deviance_file!= NULL)
initial_data->control->dev_id = TRUE;
else
initial_data->control->dev_id = -1;
if(avg_deviance_file!= NULL)
initial_data->control->avg_dev_id = TRUE;
else
initial_data->control->avg_dev_id = -1;
if(scale_file!= NULL)
initial_data->control->scale_id = TRUE;
else
initial_data->control->scale_id = -1;
if(error_file != NULL) {
initial_data->control->err_id = TRUE;
}
else {
initial_data->control->err_id = -1;
}
initial_data->control->maxVoxelSdBasedValue = maxVoxelSdBasedValue;
initial_data->control->is_fwhm = is_fwhm;
initial_data->control->use_mu_switch = use_mu_switch;
initial_data->control->max_iter = max_iter;
initial_data->control->tol = tol;
if(scale_par >= 0.0) {
initial_data->scale_est = scale_par;
initial_data->control->is_scale = TRUE;
}
else
initial_data->control->is_scale = FALSE;
initial_data->control->est_scale = est_scale;
if(chi_sq_nu > 0) {
initial_data->family = CHI_SQ;
initial_data->scale_est = 2.0 / chi_sq_nu;
}
initial_data->control->eps_sing = eps_sing;
initial_data->control->binomial_n = binomial_n;
if ((initial_data->contrast->num_contrasts == 0) &&
(initial_data->control->dev_id == -1) &&
(initial_data->control->err_id == -1) &&
(initial_data->control->scale_id == -1) &&
(lambda_buffer == NULL))
initial_data->control->do_fit = FALSE;
else
initial_data->control->do_fit = TRUE;
glm_obj = GI_MALLOC(sizeof(Glm_Object));
initialize_glm(glm_obj, initial_data);
program_data = GI_MALLOC(sizeof(*program_data));
if (lambda_buffer != NULL) {
get_fwhm_filenames(lambda_buffer, glm_obj);
if (glm_obj->control->do_fit == TRUE) {
if (glm_obj->family == GAUSSIAN)
if (lambda_buffer->fwhm_gaussian != NULL)
lambda_buffer->fwhm_gaussian->outfile = fwhm_gaussian_file;
if(glm_obj->family != GAUSSIAN) {
if(glm_obj->deg_free > 1) {
for(itest=0; itest<glm_obj->contrast->num_test; itest++) {
fwhm_general = lambda_buffer->fwhm_general[itest];
fwhm_general->constant = 1.0;
}
if (lambda_buffer->fwhm_simple != NULL) {
lambda_buffer->fwhm_simple->constant = 1.0;
}
}
else {
fprintf(stderr,"\nError: to calculate FWHM degrees of freedom must be at least 2.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
else if (lambda_buffer->fwhm_gaussian != NULL){
if(glm_obj->deg_free > 1) {
lambda_buffer->fwhm_gaussian->constant = 1.0;
}
else {
fprintf(stderr,"\nError: to calculate FWHM degrees of freedom must be at least 2.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
if(lambda_buffer->is_avg == TRUE) {
if(glm_obj->contrast->avg_array != NULL)
if (lambda_buffer->fwhm_avg != NULL)
lambda_buffer->fwhm_avg->outfile = fwhm_avg_file;
if(glm_obj->response->num_rows > 2) {
lambda_buffer->fwhm_avg->constant = 1.0;
}
else {
fprintf(stderr,"\nError: to calculate FWHM degrees of freedom must be at least 2.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
glm_obj->family_names = family_names;
glm_obj->link_names = link_names;
glm_obj->variance_function_names = variance_function_names;
glm_obj->corr_struct_names = corr_struct_names;
glm_obj->deviance_file = deviance_file;
if (glm_obj->contrast->avg_array != NULL)
glm_obj->contrast->avg_array->deviance_file = avg_deviance_file;
glm_obj->error_file = error_file;
/*Calculate x_transpsose_x_inv to be used for Gaussian */
if ((glm_obj->corr_struct == DIAGONAL_CORR) &&
(glm_obj->control->do_fit == TRUE)) {
switch(glm_obj->family) {
case GAUSSIAN:
tmp = create_matrix(design_matrix->num_columns,
design_matrix->num_columns);
multiply_matrix_sym(tmp, design_matrix, design_matrix, MATRIX_LEFT);
if (invert_matrix_sym(glm_obj->info_inv, tmp) != TRUE) {
fprintf(stderr,"\ndesign matrix:");
print_matrix(stderr, design_matrix);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
for(i=0; i<glm_obj->contrast->num_contrasts; i++) {
cur_matrix = glm_obj->contrast->contrast_matrix_array[i];
if (cur_matrix->contrast_matrix != NULL) {
multiply_matrix(glm_obj->tmp_con[i][1], glm_obj->info_inv,
cur_matrix->contrast_matrix, MATRIX_RIGHT);
multiply_matrix_sym(glm_obj->tmp_con[i][2],
cur_matrix->contrast_matrix,
glm_obj->tmp_con[i][1], MATRIX_NEITHER);
invert_matrix_sym(glm_obj->tmp_con[i][3],
glm_obj->tmp_con[i][2]);
}
}
multiply_matrix(glm_obj->lm_matrix, glm_obj->info_inv,
glm_obj->design_matrix, MATRIX_RIGHT);
delete_matrix(tmp);
break;
default:
break;
}
}
/* Checks to see if there is a need to create a temporary count file */
glm_obj->control->count_id = -1;
if(glm_obj->contrast->num_avg > 0)
glm_obj->contrast->avg_array->num_pool = 0;
for(i=0; i<glm_obj->contrast->num_avg; i++) {
cur_avg = glm_obj->contrast->avg_array->avg_info[i];
if (cur_avg->stdev_mode == POOLED_SD) {
glm_obj->control->count_id = TRUE;
glm_obj->contrast->avg_array->num_pool++;
}
}
if (glm_obj->control->deg_free_id > 0)
glm_obj->control->count_id = -1;
if(glm_obj->control->count_id == TRUE) {
tmp_count = tempnam(NULL, "glim");
fprintf(tmpfile_list, "\n%s", tmp_count);
num_tmpfiles++;
}
else
tmp_count = NULL;
/* Changing the value of num_test to make sure it's OK for gaussian */
if ((glm_obj->family == GAUSSIAN) && (lambda_buffer != NULL)) {
lambda_buffer->num_test = 1;
}
num_input_files = (design_matrix->num_rows + glm_obj->control->pool_id +
glm_obj->control->search_id);
if(lambda_buffer != NULL) {
num_output_files = (glm_obj->contrast->num_contrasts +
(glm_obj->control->avg_dev_id > 0) +
(glm_obj->control->dev_id > 0) +
(glm_obj->control->scale_id > 0) +
(glm_obj->control->err_id > 0) +
(glm_obj->control->count_id > 0) +
(lambda_buffer->num_test *
(!(lambda_buffer->is_gaussian))) +
glm_obj->contrast->num_avg +
((lambda_buffer->fwhm_gaussian != NULL) &&
(fwhm_gaussian_file != NULL)) +
((lambda_buffer->fwhm_avg != NULL) &&
(fwhm_avg_file != NULL)) +
((lambda_buffer->fwhm_simple != NULL) &&
(fwhm_simple_file != NULL)) +
(glm_obj->control->deg_free_id > 0));
}
else {
num_output_files = (glm_obj->contrast->num_contrasts +
(glm_obj->control->avg_dev_id > 0) +
(glm_obj->control->dev_id > 0) +
(glm_obj->control->scale_id > 0) +
(glm_obj->control->err_id > 0) +
(glm_obj->control->count_id > 0)+
(glm_obj->control->deg_free_id > 0) +
glm_obj->contrast->num_avg);
}
/* Set pool_id to its index in input file list */
i = design_matrix->num_rows;
if(glm_obj->control->pool_id == TRUE) {
glm_obj->control->pool_id = i;
i++;
}
else
glm_obj->control->pool_id = -1;
if(glm_obj->control->search_id == TRUE) {
glm_obj->control->search_id = i;
i++;
}
else
glm_obj->control->search_id = -1;
/* Set up file lists checking to see if to be removed */
if (glm_obj->control->pool_id > 0)
input_files[glm_obj->control->pool_id] = pool_mask;
if (glm_obj->control->search_id > 0) {
input_files[glm_obj->control->search_id] = search_mask;
glm_obj->search_mask = search_mask;
}
else
glm_obj->search_mask = NULL;
if (num_output_files <= 0) {
fprintf(stderr,"\nError: no output files specified.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
output_files = GI_MALLOC(num_output_files * sizeof(*output_files));
for(i=0; i<glm_obj->contrast->num_contrasts; i++) {
cur_matrix = glm_obj->contrast->contrast_matrix_array[i];
if(cur_matrix->stdev_mode == VOXEL_SD)
output_files[cur_matrix->out_id] = cur_matrix->outfile;
else if(cur_matrix->stdev_mode == POOLED_SD) {
if(cur_matrix->tmpfile != NULL) {
output_files[cur_matrix->out_id] = cur_matrix->tmpfile;
}
else {
fprintf(stderr,"\nError: no temporary file name for outfile %s.\n",
cur_matrix->outfile);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
for(i=0; i<glm_obj->contrast->num_avg; i++) {
cur_avg = glm_obj->contrast->avg_array->avg_info[i];
if(cur_avg->stdev_mode == VOXEL_SD)
output_files[cur_avg->out_id] = cur_avg->outfile;
else if(cur_avg->stdev_mode == POOLED_SD) {
if(cur_avg->tmpfile != NULL) {
output_files[cur_avg->out_id] = cur_avg->tmpfile;
}
else {
fprintf(stderr,"\nError: no temporary file name for outfile %s.\n",
cur_avg->outfile);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
/* Check to see that all output files have different names and whether */
/* pooling mask has been given unnecessarily. */
pool_test = TRUE;
for(i=0; i<glm_obj->contrast->num_contrasts; i++) {
cur_matrix = glm_obj->contrast->contrast_matrix_array[i];
if(cur_matrix->stdev_mode != VOXEL_SD)
pool_test = FALSE;
for(j=i+1; j<glm_obj->contrast->num_contrasts; j++) {
cur_matrix2 = glm_obj->contrast->contrast_matrix_array[j];
if(strcmp(cur_matrix->outfile, cur_matrix2->outfile) == 0) {
fprintf(stderr,"\nError: output files %d and %d have the same filename.\n", cur_matrix->out_id+1, cur_matrix2->out_id+1);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
for(i=0; i<glm_obj->contrast->num_avg; i++) {
cur_avg = glm_obj->contrast->avg_array->avg_info[i];
if(cur_avg->stdev_mode != VOXEL_SD) {
pool_test = FALSE;
}
if(glm_obj->contrast->num_avg > 1) {
for(j=i+1; j<glm_obj->contrast->num_avg; j++) {
cur_avg2 = glm_obj->contrast->avg_array->avg_info[j];
if(strcmp(cur_avg->outfile, cur_avg2->outfile) == 0) {
fprintf(stderr,"\nError: output files %d and %d have the same filename.\n", cur_avg->out_id+1, cur_avg2->out_id+1);
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
}
if((pool_test == TRUE) && (glm_obj->control->pool_id > 0)) {
fprintf(stderr,"\nError: pooling mask file given when no output files are pooled stdev type.\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
i = glm_obj->contrast->num_contrasts + glm_obj->contrast->num_avg;
if (glm_obj->control->dev_id == TRUE) {
output_files[i] = deviance_file;
glm_obj->control->dev_id = i;
i++;
}
if(glm_obj->control->deg_free_id == TRUE) {
output_files[i] = deg_free_file;
glm_obj->control->deg_free_id = i;
i++;
}
else
glm_obj->control->deg_free_id = -1;
if (glm_obj->control->avg_dev_id == TRUE) {
output_files[i] = avg_deviance_file;
glm_obj->control->avg_dev_id = i;
i++;
}
if (glm_obj->control->scale_id == TRUE) {
output_files[i] = scale_file;
glm_obj->control->scale_id = i;
glm_obj->scale_file = scale_file;
i++;
}
if (glm_obj->control->err_id == TRUE) {
output_files[i] = error_file;
glm_obj->control->err_id = i;
i++;
}
if (glm_obj->control->count_id == TRUE) {
output_files[i] = tmp_count;
glm_obj->control->count_id = i;
i++;
}
if ((glm_obj->control->is_fwhm == TRUE) && (glm_obj->family == GAUSSIAN)){
if(lambda_buffer->fwhm_gaussian != NULL) {
if(lambda_buffer->fwhm_gaussian->outfile != NULL) {
output_files[i] = lambda_buffer->fwhm_gaussian->outfile;
lambda_buffer->fwhm_gaussian->out_id = i;
i++;
}
}
}
if (lambda_buffer != NULL) {
if (glm_obj->contrast->avg_array != NULL) {
if (lambda_buffer->fwhm_avg->outfile != NULL) {
output_files[i] = lambda_buffer->fwhm_avg->outfile;
lambda_buffer->fwhm_avg->out_id = i;
i++;
}
}
if (lambda_buffer->fwhm_simple != NULL) {
if (lambda_buffer->fwhm_simple->outfile != NULL) {
output_files[i] = lambda_buffer->fwhm_simple->outfile;
lambda_buffer->fwhm_simple->out_id = i;
i++;
}
}
}
if ((glm_obj->control->is_fwhm == TRUE) && (glm_obj->family != GAUSSIAN)){
itest = 0;
for(j=i; j<i+lambda_buffer->num_test; j++) {
lambda_buffer->fwhm_general[itest]->out_id = j;
output_files[j] = lambda_buffer->fwhm_general[itest]->outfile;
itest++;
}
}
/* Check to see if pooling wanted but no pool file */
if (glm_obj->control->pool_id < 0) {
for(j=0; j<glm_obj->contrast->num_contrasts; j++) {
cur_matrix = glm_obj->contrast->contrast_matrix_array[j];
if(cur_matrix->stdev_mode == POOLED_SD) {
fprintf(stderr,"\nError: mask file needed in order to pool_sd.\n");
fprintf(stderr,"\nEnter mask file as -pool_sd <mask.mnc> on command line.\n\n");
delete_tmpfiles(&tmpfile_list);
exit(EXIT_FAILURE);
}
}
}
program_data = GI_MALLOC(sizeof(*program_data));
program_data->glm_obj = glm_obj;
program_data->lambda_buffer = lambda_buffer;
/* Set up loop options */
loop_options = create_loop_options();
set_loop_verbose(loop_options, verbose);
set_loop_clobber(loop_options, clobber);
set_loop_datatype(loop_options, datatype, is_signed,
valid_range[0], valid_range[1]);
set_loop_copy_all_header(loop_options, copy_all_header);
set_loop_buffer_size(loop_options, (long) 1024 * max_buffer_size_in_kb);
set_loop_check_dim_info(loop_options, check_dim_info);
/* Do voxel loop */
voxel_loop(num_input_files, input_files, num_output_files, output_files,
arg_string, loop_options,
voxel_function_glm, (void *) program_data);
/* Free loop options for voxel_function_pool*/
free_loop_options(loop_options);
if (glm_obj->control->pool_id > 0) {
if(glm_obj->num_pool > 0)
glm_obj->pooled_dev = glm_obj->pooled_dev / glm_obj->num_pool;