Hugging Face's logo

aehrc
/
Synthrad2025_task1

Model card Files
xet
Community
FelixzeroSun commited on
Commit
36fcd36
·
verified ·
1 Parent(s): 377fd15

Delete Dataset542

Browse files
Files changed (23) hide show
  1. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/dataset.json +0 -14
  2. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/dataset_fingerprint.json +0 -3268
  3. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_final.pth +0 -3
  4. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json +0 -58
  5. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png +0 -3
  6. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/validation/summary.json +0 -52
  7. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/checkpoint_final.pth +0 -3
  8. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/debug.json +0 -58
  9. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/progress.png +0 -3
  10. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/validation/summary.json +0 -52
  11. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/checkpoint_final.pth +0 -3
  12. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/debug.json +0 -58
  13. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/progress.png +0 -3
  14. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/validation/summary.json +0 -52
  15. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/checkpoint_final.pth +0 -3
  16. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/debug.json +0 -58
  17. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/progress.png +0 -3
  18. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/validation/summary.json +0 -52
  19. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/checkpoint_final.pth +0 -3
  20. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/debug.json +0 -58
  21. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/progress.png +0 -3
  22. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/validation/summary.json +0 -52
  23. Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/plans.json +0 -356
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/dataset.json DELETED
@@ -1,14 +0,0 @@
1
- {
2
- "labels": {
3
- "label_001": "1",
4
- "background": 0
5
- },
6
- "channel_names": {
7
- "0": "CT_zscore_synthrad"
8
- },
9
- "numTraining": 325,
10
- "file_ending": ".nii.gz",
11
- "test_set": [
12
- "case_0000"
13
- ]
14
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/dataset_fingerprint.json DELETED
@@ -1,3268 +0,0 @@
1
- {
2
- "foreground_intensity_properties_per_channel": {
3
- "0": {
4
- "max": 3071.0,
5
- "mean": -215.1082763671875,
6
- "median": -133.0,
7
- "min": -1024.0,
8
- "percentile_00_5": -1024.0,
9
- "percentile_99_5": 1345.0,
10
- "std": 459.63665771484375
11
- }
12
- },
13
- "median_relative_size_after_cropping": 1.0,
14
- "shapes_after_crop": [
15
- [
16
- 88,
17
- 307,
18
- 309
19
- ],
20
- [
21
- 83,
22
- 307,
23
- 311
24
- ],
25
- [
26
- 88,
27
- 309,
28
- 308
29
- ],
30
- [
31
- 79,
32
- 313,
33
- 314
34
- ],
35
- [
36
- 87,
37
- 309,
38
- 310
39
- ],
40
- [
41
- 88,
42
- 308,
43
- 308
44
- ],
45
- [
46
- 86,
47
- 309,
48
- 312
49
- ],
50
- [
51
- 88,
52
- 307,
53
- 308
54
- ],
55
- [
56
- 87,
57
- 309,
58
- 309
59
- ],
60
- [
61
- 87,
62
- 308,
63
- 309
64
- ],
65
- [
66
- 88,
67
- 309,
68
- 308
69
- ],
70
- [
71
- 88,
72
- 308,
73
- 307
74
- ],
75
- [
76
- 86,
77
- 313,
78
- 313
79
- ],
80
- [
81
- 82,
82
- 309,
83
- 309
84
- ],
85
- [
86
- 86,
87
- 308,
88
- 313
89
- ],
90
- [
91
- 86,
92
- 312,
93
- 307
94
- ],
95
- [
96
- 87,
97
- 307,
98
- 309
99
- ],
100
- [
101
- 87,
102
- 309,
103
- 309
104
- ],
105
- [
106
- 87,
107
- 307,
108
- 308
109
- ],
110
- [
111
- 87,
112
- 308,
113
- 309
114
- ],
115
- [
116
- 86,
117
- 309,
118
- 314
119
- ],
120
- [
121
- 88,
122
- 308,
123
- 310
124
- ],
125
- [
126
- 88,
127
- 307,
128
- 308
129
- ],
130
- [
131
- 87,
132
- 308,
133
- 307
134
- ],
135
- [
136
- 87,
137
- 311,
138
- 314
139
- ],
140
- [
141
- 87,
142
- 309,
143
- 309
144
- ],
145
- [
146
- 87,
147
- 307,
148
- 306
149
- ],
150
- [
151
- 86,
152
- 308,
153
- 310
154
- ],
155
- [
156
- 87,
157
- 307,
158
- 312
159
- ],
160
- [
161
- 87,
162
- 308,
163
- 311
164
- ],
165
- [
166
- 87,
167
- 307,
168
- 307
169
- ],
170
- [
171
- 51,
172
- 282,
173
- 311
174
- ],
175
- [
176
- 87,
177
- 308,
178
- 310
179
- ],
180
- [
181
- 88,
182
- 310,
183
- 309
184
- ],
185
- [
186
- 87,
187
- 310,
188
- 308
189
- ],
190
- [
191
- 88,
192
- 308,
193
- 308
194
- ],
195
- [
196
- 88,
197
- 308,
198
- 308
199
- ],
200
- [
201
- 87,
202
- 308,
203
- 309
204
- ],
205
- [
206
- 87,
207
- 308,
208
- 309
209
- ],
210
- [
211
- 87,
212
- 309,
213
- 310
214
- ],
215
- [
216
- 87,
217
- 308,
218
- 312
219
- ],
220
- [
221
- 82,
222
- 309,
223
- 310
224
- ],
225
- [
226
- 88,
227
- 310,
228
- 308
229
- ],
230
- [
231
- 88,
232
- 311,
233
- 308
234
- ],
235
- [
236
- 88,
237
- 309,
238
- 309
239
- ],
240
- [
241
- 87,
242
- 308,
243
- 312
244
- ],
245
- [
246
- 88,
247
- 309,
248
- 310
249
- ],
250
- [
251
- 86,
252
- 310,
253
- 312
254
- ],
255
- [
256
- 87,
257
- 310,
258
- 309
259
- ],
260
- [
261
- 87,
262
- 309,
263
- 310
264
- ],
265
- [
266
- 87,
267
- 307,
268
- 308
269
- ],
270
- [
271
- 87,
272
- 308,
273
- 310
274
- ],
275
- [
276
- 82,
277
- 288,
278
- 273
279
- ],
280
- [
281
- 88,
282
- 309,
283
- 309
284
- ],
285
- [
286
- 88,
287
- 307,
288
- 307
289
- ],
290
- [
291
- 87,
292
- 310,
293
- 307
294
- ],
295
- [
296
- 86,
297
- 311,
298
- 309
299
- ],
300
- [
301
- 87,
302
- 308,
303
- 311
304
- ],
305
- [
306
- 87,
307
- 309,
308
- 309
309
- ],
310
- [
311
- 87,
312
- 310,
313
- 308
314
- ],
315
- [
316
- 88,
317
- 225,
318
- 308
319
- ],
320
- [
321
- 87,
322
- 308,
323
- 307
324
- ],
325
- [
326
- 81,
327
- 311,
328
- 308
329
- ],
330
- [
331
- 87,
332
- 310,
333
- 310
334
- ],
335
- [
336
- 87,
337
- 309,
338
- 307
339
- ],
340
- [
341
- 88,
342
- 310,
343
- 308
344
- ],
345
- [
346
- 87,
347
- 313,
348
- 311
349
- ],
350
- [
351
- 88,
352
- 309,
353
- 309
354
- ],
355
- [
356
- 88,
357
- 309,
358
- 307
359
- ],
360
- [
361
- 88,
362
- 309,
363
- 308
364
- ],
365
- [
366
- 88,
367
- 309,
368
- 309
369
- ],
370
- [
371
- 87,
372
- 308,
373
- 312
374
- ],
375
- [
376
- 87,
377
- 308,
378
- 310
379
- ],
380
- [
381
- 88,
382
- 311,
383
- 309
384
- ],
385
- [
386
- 88,
387
- 308,
388
- 308
389
- ],
390
- [
391
- 88,
392
- 308,
393
- 308
394
- ],
395
- [
396
- 49,
397
- 307,
398
- 307
399
- ],
400
- [
401
- 88,
402
- 308,
403
- 309
404
- ],
405
- [
406
- 88,
407
- 310,
408
- 307
409
- ],
410
- [
411
- 86,
412
- 315,
413
- 315
414
- ],
415
- [
416
- 88,
417
- 306,
418
- 307
419
- ],
420
- [
421
- 87,
422
- 310,
423
- 311
424
- ],
425
- [
426
- 87,
427
- 311,
428
- 308
429
- ],
430
- [
431
- 88,
432
- 309,
433
- 308
434
- ],
435
- [
436
- 87,
437
- 310,
438
- 309
439
- ],
440
- [
441
- 88,
442
- 309,
443
- 309
444
- ],
445
- [
446
- 87,
447
- 312,
448
- 309
449
- ],
450
- [
451
- 87,
452
- 310,
453
- 311
454
- ],
455
- [
456
- 88,
457
- 310,
458
- 311
459
- ],
460
- [
461
- 88,
462
- 308,
463
- 309
464
- ],
465
- [
466
- 87,
467
- 309,
468
- 310
469
- ],
470
- [
471
- 87,
472
- 310,
473
- 311
474
- ],
475
- [
476
- 87,
477
- 312,
478
- 311
479
- ],
480
- [
481
- 88,
482
- 309,
483
- 311
484
- ],
485
- [
486
- 88,
487
- 308,
488
- 310
489
- ],
490
- [
491
- 88,
492
- 310,
493
- 309
494
- ],
495
- [
496
- 87,
497
- 308,
498
- 310
499
- ],
500
- [
501
- 87,
502
- 311,
503
- 308
504
- ],
505
- [
506
- 88,
507
- 309,
508
- 308
509
- ],
510
- [
511
- 87,
512
- 311,
513
- 311
514
- ],
515
- [
516
- 87,
517
- 307,
518
- 309
519
- ],
520
- [
521
- 87,
522
- 310,
523
- 311
524
- ],
525
- [
526
- 88,
527
- 309,
528
- 309
529
- ],
530
- [
531
- 82,
532
- 309,
533
- 308
534
- ],
535
- [
536
- 88,
537
- 309,
538
- 309
539
- ],
540
- [
541
- 87,
542
- 311,
543
- 308
544
- ],
545
- [
546
- 87,
547
- 310,
548
- 313
549
- ],
550
- [
551
- 88,
552
- 310,
553
- 309
554
- ],
555
- [
556
- 87,
557
- 309,
558
- 311
559
- ],
560
- [
561
- 88,
562
- 310,
563
- 309
564
- ],
565
- [
566
- 88,
567
- 308,
568
- 308
569
- ],
570
- [
571
- 87,
572
- 310,
573
- 309
574
- ],
575
- [
576
- 87,
577
- 309,
578
- 310
579
- ],
580
- [
581
- 87,
582
- 312,
583
- 312
584
- ],
585
- [
586
- 87,
587
- 309,
588
- 311
589
- ],
590
- [
591
- 87,
592
- 312,
593
- 311
594
- ],
595
- [
596
- 87,
597
- 312,
598
- 310
599
- ],
600
- [
601
- 87,
602
- 307,
603
- 309
604
- ],
605
- [
606
- 87,
607
- 312,
608
- 308
609
- ],
610
- [
611
- 88,
612
- 307,
613
- 309
614
- ],
615
- [
616
- 87,
617
- 308,
618
- 312
619
- ],
620
- [
621
- 87,
622
- 311,
623
- 307
624
- ],
625
- [
626
- 87,
627
- 312,
628
- 311
629
- ],
630
- [
631
- 81,
632
- 310,
633
- 311
634
- ],
635
- [
636
- 87,
637
- 308,
638
- 312
639
- ],
640
- [
641
- 88,
642
- 310,
643
- 308
644
- ],
645
- [
646
- 87,
647
- 311,
648
- 309
649
- ],
650
- [
651
- 87,
652
- 309,
653
- 310
654
- ],
655
- [
656
- 86,
657
- 311,
658
- 317
659
- ],
660
- [
661
- 88,
662
- 308,
663
- 308
664
- ],
665
- [
666
- 88,
667
- 311,
668
- 309
669
- ],
670
- [
671
- 51,
672
- 311,
673
- 311
674
- ],
675
- [
676
- 87,
677
- 311,
678
- 309
679
- ],
680
- [
681
- 86,
682
- 310,
683
- 307
684
- ],
685
- [
686
- 88,
687
- 311,
688
- 308
689
- ],
690
- [
691
- 88,
692
- 309,
693
- 309
694
- ],
695
- [
696
- 87,
697
- 311,
698
- 312
699
- ],
700
- [
701
- 87,
702
- 310,
703
- 307
704
- ],
705
- [
706
- 86,
707
- 309,
708
- 311
709
- ],
710
- [
711
- 88,
712
- 310,
713
- 311
714
- ],
715
- [
716
- 87,
717
- 312,
718
- 308
719
- ],
720
- [
721
- 86,
722
- 308,
723
- 311
724
- ],
725
- [
726
- 87,
727
- 311,
728
- 312
729
- ],
730
- [
731
- 85,
732
- 313,
733
- 314
734
- ],
735
- [
736
- 86,
737
- 311,
738
- 311
739
- ],
740
- [
741
- 88,
742
- 308,
743
- 309
744
- ],
745
- [
746
- 86,
747
- 314,
748
- 313
749
- ],
750
- [
751
- 87,
752
- 312,
753
- 312
754
- ],
755
- [
756
- 87,
757
- 314,
758
- 311
759
- ],
760
- [
761
- 86,
762
- 308,
763
- 312
764
- ],
765
- [
766
- 87,
767
- 308,
768
- 312
769
- ],
770
- [
771
- 88,
772
- 313,
773
- 308
774
- ],
775
- [
776
- 87,
777
- 307,
778
- 309
779
- ],
780
- [
781
- 87,
782
- 311,
783
- 308
784
- ],
785
- [
786
- 85,
787
- 317,
788
- 311
789
- ],
790
- [
791
- 87,
792
- 307,
793
- 311
794
- ],
795
- [
796
- 88,
797
- 308,
798
- 312
799
- ],
800
- [
801
- 87,
802
- 310,
803
- 310
804
- ],
805
- [
806
- 87,
807
- 311,
808
- 311
809
- ],
810
- [
811
- 52,
812
- 308,
813
- 310
814
- ],
815
- [
816
- 87,
817
- 309,
818
- 307
819
- ],
820
- [
821
- 87,
822
- 308,
823
- 311
824
- ],
825
- [
826
- 87,
827
- 308,
828
- 310
829
- ],
830
- [
831
- 87,
832
- 310,
833
- 309
834
- ],
835
- [
836
- 88,
837
- 309,
838
- 309
839
- ],
840
- [
841
- 87,
842
- 309,
843
- 309
844
- ],
845
- [
846
- 86,
847
- 310,
848
- 314
849
- ],
850
- [
851
- 88,
852
- 309,
853
- 308
854
- ],
855
- [
856
- 86,
857
- 314,
858
- 312
859
- ],
860
- [
861
- 88,
862
- 308,
863
- 309
864
- ],
865
- [
866
- 87,
867
- 309,
868
- 310
869
- ],
870
- [
871
- 87,
872
- 309,
873
- 310
874
- ],
875
- [
876
- 87,
877
- 312,
878
- 309
879
- ],
880
- [
881
- 88,
882
- 310,
883
- 310
884
- ],
885
- [
886
- 86,
887
- 313,
888
- 311
889
- ],
890
- [
891
- 87,
892
- 311,
893
- 310
894
- ],
895
- [
896
- 87,
897
- 311,
898
- 309
899
- ],
900
- [
901
- 86,
902
- 311,
903
- 312
904
- ],
905
- [
906
- 87,
907
- 311,
908
- 310
909
- ],
910
- [
911
- 87,
912
- 309,
913
- 307
914
- ],
915
- [
916
- 85,
917
- 310,
918
- 311
919
- ],
920
- [
921
- 88,
922
- 312,
923
- 308
924
- ],
925
- [
926
- 85,
927
- 293,
928
- 319
929
- ],
930
- [
931
- 86,
932
- 308,
933
- 308
934
- ],
935
- [
936
- 87,
937
- 310,
938
- 309
939
- ],
940
- [
941
- 87,
942
- 309,
943
- 310
944
- ],
945
- [
946
- 53,
947
- 307,
948
- 310
949
- ],
950
- [
951
- 87,
952
- 308,
953
- 312
954
- ],
955
- [
956
- 54,
957
- 308,
958
- 307
959
- ],
960
- [
961
- 88,
962
- 309,
963
- 307
964
- ],
965
- [
966
- 88,
967
- 308,
968
- 307
969
- ],
970
- [
971
- 87,
972
- 307,
973
- 310
974
- ],
975
- [
976
- 88,
977
- 307,
978
- 308
979
- ],
980
- [
981
- 87,
982
- 310,
983
- 307
984
- ],
985
- [
986
- 77,
987
- 311,
988
- 311
989
- ],
990
- [
991
- 99,
992
- 315,
993
- 316
994
- ],
995
- [
996
- 128,
997
- 305,
998
- 307
999
- ],
1000
- [
1001
- 129,
1002
- 300,
1003
- 305
1004
- ],
1005
- [
1006
- 128,
1007
- 303,
1008
- 305
1009
- ],
1010
- [
1011
- 121,
1012
- 299,
1013
- 300
1014
- ],
1015
- [
1016
- 128,
1017
- 304,
1018
- 301
1019
- ],
1020
- [
1021
- 70,
1022
- 300,
1023
- 298
1024
- ],
1025
- [
1026
- 70,
1027
- 300,
1028
- 303
1029
- ],
1030
- [
1031
- 71,
1032
- 304,
1033
- 305
1034
- ],
1035
- [
1036
- 71,
1037
- 302,
1038
- 303
1039
- ],
1040
- [
1041
- 72,
1042
- 302,
1043
- 299
1044
- ],
1045
- [
1046
- 70,
1047
- 300,
1048
- 301
1049
- ],
1050
- [
1051
- 99,
1052
- 308,
1053
- 312
1054
- ],
1055
- [
1056
- 70,
1057
- 303,
1058
- 299
1059
- ],
1060
- [
1061
- 71,
1062
- 300,
1063
- 300
1064
- ],
1065
- [
1066
- 70,
1067
- 301,
1068
- 299
1069
- ],
1070
- [
1071
- 72,
1072
- 307,
1073
- 306
1074
- ],
1075
- [
1076
- 70,
1077
- 304,
1078
- 304
1079
- ],
1080
- [
1081
- 121,
1082
- 303,
1083
- 306
1084
- ],
1085
- [
1086
- 119,
1087
- 303,
1088
- 303
1089
- ],
1090
- [
1091
- 122,
1092
- 308,
1093
- 301
1094
- ],
1095
- [
1096
- 130,
1097
- 310,
1098
- 301
1099
- ],
1100
- [
1101
- 71,
1102
- 300,
1103
- 300
1104
- ],
1105
- [
1106
- 128,
1107
- 301,
1108
- 301
1109
- ],
1110
- [
1111
- 130,
1112
- 392,
1113
- 400
1114
- ],
1115
- [
1116
- 73,
1117
- 302,
1118
- 302
1119
- ],
1120
- [
1121
- 70,
1122
- 301,
1123
- 302
1124
- ],
1125
- [
1126
- 70,
1127
- 301,
1128
- 301
1129
- ],
1130
- [
1131
- 71,
1132
- 299,
1133
- 301
1134
- ],
1135
- [
1136
- 71,
1137
- 304,
1138
- 303
1139
- ],
1140
- [
1141
- 71,
1142
- 302,
1143
- 298
1144
- ],
1145
- [
1146
- 70,
1147
- 299,
1148
- 300
1149
- ],
1150
- [
1151
- 130,
1152
- 371,
1153
- 394
1154
- ],
1155
- [
1156
- 123,
1157
- 285,
1158
- 306
1159
- ],
1160
- [
1161
- 70,
1162
- 300,
1163
- 299
1164
- ],
1165
- [
1166
- 70,
1167
- 301,
1168
- 299
1169
- ],
1170
- [
1171
- 116,
1172
- 302,
1173
- 302
1174
- ],
1175
- [
1176
- 129,
1177
- 299,
1178
- 302
1179
- ],
1180
- [
1181
- 127,
1182
- 304,
1183
- 303
1184
- ],
1185
- [
1186
- 115,
1187
- 311,
1188
- 311
1189
- ],
1190
- [
1191
- 71,
1192
- 299,
1193
- 302
1194
- ],
1195
- [
1196
- 70,
1197
- 300,
1198
- 301
1199
- ],
1200
- [
1201
- 72,
1202
- 304,
1203
- 300
1204
- ],
1205
- [
1206
- 72,
1207
- 305,
1208
- 300
1209
- ],
1210
- [
1211
- 99,
1212
- 290,
1213
- 313
1214
- ],
1215
- [
1216
- 70,
1217
- 300,
1218
- 299
1219
- ],
1220
- [
1221
- 111,
1222
- 302,
1223
- 303
1224
- ],
1225
- [
1226
- 128,
1227
- 300,
1228
- 298
1229
- ],
1230
- [
1231
- 122,
1232
- 305,
1233
- 302
1234
- ],
1235
- [
1236
- 70,
1237
- 300,
1238
- 299
1239
- ],
1240
- [
1241
- 129,
1242
- 303,
1243
- 301
1244
- ],
1245
- [
1246
- 121,
1247
- 310,
1248
- 305
1249
- ],
1250
- [
1251
- 122,
1252
- 301,
1253
- 302
1254
- ],
1255
- [
1256
- 110,
1257
- 306,
1258
- 298
1259
- ],
1260
- [
1261
- 119,
1262
- 310,
1263
- 308
1264
- ],
1265
- [
1266
- 121,
1267
- 302,
1268
- 304
1269
- ],
1270
- [
1271
- 116,
1272
- 304,
1273
- 303
1274
- ],
1275
- [
1276
- 124,
1277
- 308,
1278
- 313
1279
- ],
1280
- [
1281
- 72,
1282
- 304,
1283
- 303
1284
- ],
1285
- [
1286
- 69,
1287
- 298,
1288
- 299
1289
- ],
1290
- [
1291
- 71,
1292
- 299,
1293
- 301
1294
- ],
1295
- [
1296
- 122,
1297
- 303,
1298
- 299
1299
- ],
1300
- [
1301
- 70,
1302
- 301,
1303
- 300
1304
- ],
1305
- [
1306
- 131,
1307
- 313,
1308
- 310
1309
- ],
1310
- [
1311
- 70,
1312
- 302,
1313
- 301
1314
- ],
1315
- [
1316
- 72,
1317
- 299,
1318
- 298
1319
- ],
1320
- [
1321
- 66,
1322
- 496,
1323
- 498
1324
- ],
1325
- [
1326
- 71,
1327
- 297,
1328
- 298
1329
- ],
1330
- [
1331
- 74,
1332
- 305,
1333
- 304
1334
- ],
1335
- [
1336
- 72,
1337
- 298,
1338
- 297
1339
- ],
1340
- [
1341
- 71,
1342
- 297,
1343
- 297
1344
- ],
1345
- [
1346
- 71,
1347
- 299,
1348
- 297
1349
- ],
1350
- [
1351
- 73,
1352
- 298,
1353
- 299
1354
- ],
1355
- [
1356
- 71,
1357
- 298,
1358
- 297
1359
- ],
1360
- [
1361
- 68,
1362
- 457,
1363
- 498
1364
- ],
1365
- [
1366
- 71,
1367
- 298,
1368
- 297
1369
- ],
1370
- [
1371
- 71,
1372
- 298,
1373
- 296
1374
- ],
1375
- [
1376
- 72,
1377
- 297,
1378
- 298
1379
- ],
1380
- [
1381
- 72,
1382
- 297,
1383
- 301
1384
- ],
1385
- [
1386
- 66,
1387
- 482,
1388
- 498
1389
- ],
1390
- [
1391
- 67,
1392
- 496,
1393
- 498
1394
- ],
1395
- [
1396
- 70,
1397
- 500,
1398
- 498
1399
- ],
1400
- [
1401
- 71,
1402
- 298,
1403
- 296
1404
- ],
1405
- [
1406
- 71,
1407
- 296,
1408
- 295
1409
- ],
1410
- [
1411
- 66,
1412
- 497,
1413
- 497
1414
- ],
1415
- [
1416
- 73,
1417
- 300,
1418
- 299
1419
- ],
1420
- [
1421
- 71,
1422
- 297,
1423
- 297
1424
- ],
1425
- [
1426
- 71,
1427
- 296,
1428
- 297
1429
- ],
1430
- [
1431
- 69,
1432
- 480,
1433
- 499
1434
- ],
1435
- [
1436
- 71,
1437
- 298,
1438
- 298
1439
- ],
1440
- [
1441
- 71,
1442
- 297,
1443
- 297
1444
- ],
1445
- [
1446
- 73,
1447
- 297,
1448
- 303
1449
- ],
1450
- [
1451
- 73,
1452
- 299,
1453
- 299
1454
- ],
1455
- [
1456
- 72,
1457
- 299,
1458
- 298
1459
- ],
1460
- [
1461
- 71,
1462
- 297,
1463
- 295
1464
- ],
1465
- [
1466
- 71,
1467
- 299,
1468
- 299
1469
- ],
1470
- [
1471
- 71,
1472
- 298,
1473
- 296
1474
- ],
1475
- [
1476
- 70,
1477
- 493,
1478
- 499
1479
- ],
1480
- [
1481
- 71,
1482
- 296,
1483
- 297
1484
- ],
1485
- [
1486
- 68,
1487
- 497,
1488
- 500
1489
- ],
1490
- [
1491
- 66,
1492
- 496,
1493
- 496
1494
- ],
1495
- [
1496
- 71,
1497
- 484,
1498
- 501
1499
- ],
1500
- [
1501
- 74,
1502
- 297,
1503
- 301
1504
- ],
1505
- [
1506
- 73,
1507
- 297,
1508
- 297
1509
- ],
1510
- [
1511
- 73,
1512
- 300,
1513
- 298
1514
- ],
1515
- [
1516
- 72,
1517
- 298,
1518
- 298
1519
- ],
1520
- [
1521
- 71,
1522
- 499,
1523
- 470
1524
- ],
1525
- [
1526
- 71,
1527
- 297,
1528
- 298
1529
- ],
1530
- [
1531
- 72,
1532
- 498,
1533
- 503
1534
- ],
1535
- [
1536
- 71,
1537
- 297,
1538
- 296
1539
- ],
1540
- [
1541
- 71,
1542
- 300,
1543
- 298
1544
- ],
1545
- [
1546
- 71,
1547
- 298,
1548
- 296
1549
- ],
1550
- [
1551
- 67,
1552
- 497,
1553
- 497
1554
- ],
1555
- [
1556
- 71,
1557
- 501,
1558
- 499
1559
- ],
1560
- [
1561
- 69,
1562
- 498,
1563
- 501
1564
- ],
1565
- [
1566
- 70,
1567
- 297,
1568
- 297
1569
- ],
1570
- [
1571
- 66,
1572
- 498,
1573
- 477
1574
- ],
1575
- [
1576
- 71,
1577
- 500,
1578
- 500
1579
- ],
1580
- [
1581
- 68,
1582
- 483,
1583
- 499
1584
- ],
1585
- [
1586
- 66,
1587
- 497,
1588
- 498
1589
- ],
1590
- [
1591
- 76,
1592
- 306,
1593
- 301
1594
- ],
1595
- [
1596
- 73,
1597
- 299,
1598
- 298
1599
- ],
1600
- [
1601
- 71,
1602
- 296,
1603
- 298
1604
- ],
1605
- [
1606
- 70,
1607
- 499,
1608
- 498
1609
- ],
1610
- [
1611
- 70,
1612
- 499,
1613
- 499
1614
- ],
1615
- [
1616
- 72,
1617
- 498,
1618
- 504
1619
- ],
1620
- [
1621
- 71,
1622
- 490,
1623
- 502
1624
- ],
1625
- [
1626
- 71,
1627
- 501,
1628
- 500
1629
- ],
1630
- [
1631
- 71,
1632
- 501,
1633
- 499
1634
- ],
1635
- [
1636
- 71,
1637
- 499,
1638
- 501
1639
- ]
1640
- ],
1641
- "spacings": [
1642
- [
1643
- 3.0,
1644
- 1.0,
1645
- 1.0
1646
- ],
1647
- [
1648
- 3.0,
1649
- 1.0,
1650
- 1.0
1651
- ],
1652
- [
1653
- 3.0,
1654
- 1.0,
1655
- 1.0
1656
- ],
1657
- [
1658
- 3.0,
1659
- 1.0,
1660
- 1.0
1661
- ],
1662
- [
1663
- 3.0,
1664
- 1.0,
1665
- 1.0
1666
- ],
1667
- [
1668
- 3.0,
1669
- 1.0,
1670
- 1.0
1671
- ],
1672
- [
1673
- 3.0,
1674
- 1.0,
1675
- 1.0
1676
- ],
1677
- [
1678
- 3.0,
1679
- 1.0,
1680
- 1.0
1681
- ],
1682
- [
1683
- 3.0,
1684
- 1.0,
1685
- 1.0
1686
- ],
1687
- [
1688
- 3.0,
1689
- 1.0,
1690
- 1.0
1691
- ],
1692
- [
1693
- 3.0,
1694
- 1.0,
1695
- 1.0
1696
- ],
1697
- [
1698
- 3.0,
1699
- 1.0,
1700
- 1.0
1701
- ],
1702
- [
1703
- 3.0,
1704
- 1.0,
1705
- 1.0
1706
- ],
1707
- [
1708
- 3.0,
1709
- 1.0,
1710
- 1.0
1711
- ],
1712
- [
1713
- 3.0,
1714
- 1.0,
1715
- 1.0
1716
- ],
1717
- [
1718
- 3.0,
1719
- 1.0,
1720
- 1.0
1721
- ],
1722
- [
1723
- 3.0,
1724
- 1.0,
1725
- 1.0
1726
- ],
1727
- [
1728
- 3.0,
1729
- 1.0,
1730
- 1.0
1731
- ],
1732
- [
1733
- 3.0,
1734
- 1.0,
1735
- 1.0
1736
- ],
1737
- [
1738
- 3.0,
1739
- 1.0,
1740
- 1.0
1741
- ],
1742
- [
1743
- 3.0,
1744
- 1.0,
1745
- 1.0
1746
- ],
1747
- [
1748
- 3.0,
1749
- 1.0,
1750
- 1.0
1751
- ],
1752
- [
1753
- 3.0,
1754
- 1.0,
1755
- 1.0
1756
- ],
1757
- [
1758
- 3.0,
1759
- 1.0,
1760
- 1.0
1761
- ],
1762
- [
1763
- 3.0,
1764
- 1.0,
1765
- 1.0
1766
- ],
1767
- [
1768
- 3.0,
1769
- 1.0,
1770
- 1.0
1771
- ],
1772
- [
1773
- 3.0,
1774
- 1.0,
1775
- 1.0
1776
- ],
1777
- [
1778
- 3.0,
1779
- 1.0,
1780
- 1.0
1781
- ],
1782
- [
1783
- 3.0,
1784
- 1.0,
1785
- 1.0
1786
- ],
1787
- [
1788
- 3.0,
1789
- 1.0,
1790
- 1.0
1791
- ],
1792
- [
1793
- 3.0,
1794
- 1.0,
1795
- 1.0
1796
- ],
1797
- [
1798
- 3.0,
1799
- 1.0,
1800
- 1.0
1801
- ],
1802
- [
1803
- 3.0,
1804
- 1.0,
1805
- 1.0
1806
- ],
1807
- [
1808
- 3.0,
1809
- 1.0,
1810
- 1.0
1811
- ],
1812
- [
1813
- 3.0,
1814
- 1.0,
1815
- 1.0
1816
- ],
1817
- [
1818
- 3.0,
1819
- 1.0,
1820
- 1.0
1821
- ],
1822
- [
1823
- 3.0,
1824
- 1.0,
1825
- 1.0
1826
- ],
1827
- [
1828
- 3.0,
1829
- 1.0,
1830
- 1.0
1831
- ],
1832
- [
1833
- 3.0,
1834
- 1.0,
1835
- 1.0
1836
- ],
1837
- [
1838
- 3.0,
1839
- 1.0,
1840
- 1.0
1841
- ],
1842
- [
1843
- 3.0,
1844
- 1.0,
1845
- 1.0
1846
- ],
1847
- [
1848
- 3.0,
1849
- 1.0,
1850
- 1.0
1851
- ],
1852
- [
1853
- 3.0,
1854
- 1.0,
1855
- 1.0
1856
- ],
1857
- [
1858
- 3.0,
1859
- 1.0,
1860
- 1.0
1861
- ],
1862
- [
1863
- 3.0,
1864
- 1.0,
1865
- 1.0
1866
- ],
1867
- [
1868
- 3.0,
1869
- 1.0,
1870
- 1.0
1871
- ],
1872
- [
1873
- 3.0,
1874
- 1.0,
1875
- 1.0
1876
- ],
1877
- [
1878
- 3.0,
1879
- 1.0,
1880
- 1.0
1881
- ],
1882
- [
1883
- 3.0,
1884
- 1.0,
1885
- 1.0
1886
- ],
1887
- [
1888
- 3.0,
1889
- 1.0,
1890
- 1.0
1891
- ],
1892
- [
1893
- 3.0,
1894
- 1.0,
1895
- 1.0
1896
- ],
1897
- [
1898
- 3.0,
1899
- 1.0,
1900
- 1.0
1901
- ],
1902
- [
1903
- 3.0,
1904
- 1.0,
1905
- 1.0
1906
- ],
1907
- [
1908
- 3.0,
1909
- 1.0,
1910
- 1.0
1911
- ],
1912
- [
1913
- 3.0,
1914
- 1.0,
1915
- 1.0
1916
- ],
1917
- [
1918
- 3.0,
1919
- 1.0,
1920
- 1.0
1921
- ],
1922
- [
1923
- 3.0,
1924
- 1.0,
1925
- 1.0
1926
- ],
1927
- [
1928
- 3.0,
1929
- 1.0,
1930
- 1.0
1931
- ],
1932
- [
1933
- 3.0,
1934
- 1.0,
1935
- 1.0
1936
- ],
1937
- [
1938
- 3.0,
1939
- 1.0,
1940
- 1.0
1941
- ],
1942
- [
1943
- 3.0,
1944
- 1.0,
1945
- 1.0
1946
- ],
1947
- [
1948
- 3.0,
1949
- 1.0,
1950
- 1.0
1951
- ],
1952
- [
1953
- 3.0,
1954
- 1.0,
1955
- 1.0
1956
- ],
1957
- [
1958
- 3.0,
1959
- 1.0,
1960
- 1.0
1961
- ],
1962
- [
1963
- 3.0,
1964
- 1.0,
1965
- 1.0
1966
- ],
1967
- [
1968
- 3.0,
1969
- 1.0,
1970
- 1.0
1971
- ],
1972
- [
1973
- 3.0,
1974
- 1.0,
1975
- 1.0
1976
- ],
1977
- [
1978
- 3.0,
1979
- 1.0,
1980
- 1.0
1981
- ],
1982
- [
1983
- 3.0,
1984
- 1.0,
1985
- 1.0
1986
- ],
1987
- [
1988
- 3.0,
1989
- 1.0,
1990
- 1.0
1991
- ],
1992
- [
1993
- 3.0,
1994
- 1.0,
1995
- 1.0
1996
- ],
1997
- [
1998
- 3.0,
1999
- 1.0,
2000
- 1.0
2001
- ],
2002
- [
2003
- 3.0,
2004
- 1.0,
2005
- 1.0
2006
- ],
2007
- [
2008
- 3.0,
2009
- 1.0,
2010
- 1.0
2011
- ],
2012
- [
2013
- 3.0,
2014
- 1.0,
2015
- 1.0
2016
- ],
2017
- [
2018
- 3.0,
2019
- 1.0,
2020
- 1.0
2021
- ],
2022
- [
2023
- 3.0,
2024
- 1.0,
2025
- 1.0
2026
- ],
2027
- [
2028
- 3.0,
2029
- 1.0,
2030
- 1.0
2031
- ],
2032
- [
2033
- 3.0,
2034
- 1.0,
2035
- 1.0
2036
- ],
2037
- [
2038
- 3.0,
2039
- 1.0,
2040
- 1.0
2041
- ],
2042
- [
2043
- 3.0,
2044
- 1.0,
2045
- 1.0
2046
- ],
2047
- [
2048
- 3.0,
2049
- 1.0,
2050
- 1.0
2051
- ],
2052
- [
2053
- 3.0,
2054
- 1.0,
2055
- 1.0
2056
- ],
2057
- [
2058
- 3.0,
2059
- 1.0,
2060
- 1.0
2061
- ],
2062
- [
2063
- 3.0,
2064
- 1.0,
2065
- 1.0
2066
- ],
2067
- [
2068
- 3.0,
2069
- 1.0,
2070
- 1.0
2071
- ],
2072
- [
2073
- 3.0,
2074
- 1.0,
2075
- 1.0
2076
- ],
2077
- [
2078
- 3.0,
2079
- 1.0,
2080
- 1.0
2081
- ],
2082
- [
2083
- 3.0,
2084
- 1.0,
2085
- 1.0
2086
- ],
2087
- [
2088
- 3.0,
2089
- 1.0,
2090
- 1.0
2091
- ],
2092
- [
2093
- 3.0,
2094
- 1.0,
2095
- 1.0
2096
- ],
2097
- [
2098
- 3.0,
2099
- 1.0,
2100
- 1.0
2101
- ],
2102
- [
2103
- 3.0,
2104
- 1.0,
2105
- 1.0
2106
- ],
2107
- [
2108
- 3.0,
2109
- 1.0,
2110
- 1.0
2111
- ],
2112
- [
2113
- 3.0,
2114
- 1.0,
2115
- 1.0
2116
- ],
2117
- [
2118
- 3.0,
2119
- 1.0,
2120
- 1.0
2121
- ],
2122
- [
2123
- 3.0,
2124
- 1.0,
2125
- 1.0
2126
- ],
2127
- [
2128
- 3.0,
2129
- 1.0,
2130
- 1.0
2131
- ],
2132
- [
2133
- 3.0,
2134
- 1.0,
2135
- 1.0
2136
- ],
2137
- [
2138
- 3.0,
2139
- 1.0,
2140
- 1.0
2141
- ],
2142
- [
2143
- 3.0,
2144
- 1.0,
2145
- 1.0
2146
- ],
2147
- [
2148
- 3.0,
2149
- 1.0,
2150
- 1.0
2151
- ],
2152
- [
2153
- 3.0,
2154
- 1.0,
2155
- 1.0
2156
- ],
2157
- [
2158
- 3.0,
2159
- 1.0,
2160
- 1.0
2161
- ],
2162
- [
2163
- 3.0,
2164
- 1.0,
2165
- 1.0
2166
- ],
2167
- [
2168
- 3.0,
2169
- 1.0,
2170
- 1.0
2171
- ],
2172
- [
2173
- 3.0,
2174
- 1.0,
2175
- 1.0
2176
- ],
2177
- [
2178
- 3.0,
2179
- 1.0,
2180
- 1.0
2181
- ],
2182
- [
2183
- 3.0,
2184
- 1.0,
2185
- 1.0
2186
- ],
2187
- [
2188
- 3.0,
2189
- 1.0,
2190
- 1.0
2191
- ],
2192
- [
2193
- 3.0,
2194
- 1.0,
2195
- 1.0
2196
- ],
2197
- [
2198
- 3.0,
2199
- 1.0,
2200
- 1.0
2201
- ],
2202
- [
2203
- 3.0,
2204
- 1.0,
2205
- 1.0
2206
- ],
2207
- [
2208
- 3.0,
2209
- 1.0,
2210
- 1.0
2211
- ],
2212
- [
2213
- 3.0,
2214
- 1.0,
2215
- 1.0
2216
- ],
2217
- [
2218
- 3.0,
2219
- 1.0,
2220
- 1.0
2221
- ],
2222
- [
2223
- 3.0,
2224
- 1.0,
2225
- 1.0
2226
- ],
2227
- [
2228
- 3.0,
2229
- 1.0,
2230
- 1.0
2231
- ],
2232
- [
2233
- 3.0,
2234
- 1.0,
2235
- 1.0
2236
- ],
2237
- [
2238
- 3.0,
2239
- 1.0,
2240
- 1.0
2241
- ],
2242
- [
2243
- 3.0,
2244
- 1.0,
2245
- 1.0
2246
- ],
2247
- [
2248
- 3.0,
2249
- 1.0,
2250
- 1.0
2251
- ],
2252
- [
2253
- 3.0,
2254
- 1.0,
2255
- 1.0
2256
- ],
2257
- [
2258
- 3.0,
2259
- 1.0,
2260
- 1.0
2261
- ],
2262
- [
2263
- 3.0,
2264
- 1.0,
2265
- 1.0
2266
- ],
2267
- [
2268
- 3.0,
2269
- 1.0,
2270
- 1.0
2271
- ],
2272
- [
2273
- 3.0,
2274
- 1.0,
2275
- 1.0
2276
- ],
2277
- [
2278
- 3.0,
2279
- 1.0,
2280
- 1.0
2281
- ],
2282
- [
2283
- 3.0,
2284
- 1.0,
2285
- 1.0
2286
- ],
2287
- [
2288
- 3.0,
2289
- 1.0,
2290
- 1.0
2291
- ],
2292
- [
2293
- 3.0,
2294
- 1.0,
2295
- 1.0
2296
- ],
2297
- [
2298
- 3.0,
2299
- 1.0,
2300
- 1.0
2301
- ],
2302
- [
2303
- 3.0,
2304
- 1.0,
2305
- 1.0
2306
- ],
2307
- [
2308
- 3.0,
2309
- 1.0,
2310
- 1.0
2311
- ],
2312
- [
2313
- 3.0,
2314
- 1.0,
2315
- 1.0
2316
- ],
2317
- [
2318
- 3.0,
2319
- 1.0,
2320
- 1.0
2321
- ],
2322
- [
2323
- 3.0,
2324
- 1.0,
2325
- 1.0
2326
- ],
2327
- [
2328
- 3.0,
2329
- 1.0,
2330
- 1.0
2331
- ],
2332
- [
2333
- 3.0,
2334
- 1.0,
2335
- 1.0
2336
- ],
2337
- [
2338
- 3.0,
2339
- 1.0,
2340
- 1.0
2341
- ],
2342
- [
2343
- 3.0,
2344
- 1.0,
2345
- 1.0
2346
- ],
2347
- [
2348
- 3.0,
2349
- 1.0,
2350
- 1.0
2351
- ],
2352
- [
2353
- 3.0,
2354
- 1.0,
2355
- 1.0
2356
- ],
2357
- [
2358
- 3.0,
2359
- 1.0,
2360
- 1.0
2361
- ],
2362
- [
2363
- 3.0,
2364
- 1.0,
2365
- 1.0
2366
- ],
2367
- [
2368
- 3.0,
2369
- 1.0,
2370
- 1.0
2371
- ],
2372
- [
2373
- 3.0,
2374
- 1.0,
2375
- 1.0
2376
- ],
2377
- [
2378
- 3.0,
2379
- 1.0,
2380
- 1.0
2381
- ],
2382
- [
2383
- 3.0,
2384
- 1.0,
2385
- 1.0
2386
- ],
2387
- [
2388
- 3.0,
2389
- 1.0,
2390
- 1.0
2391
- ],
2392
- [
2393
- 3.0,
2394
- 1.0,
2395
- 1.0
2396
- ],
2397
- [
2398
- 3.0,
2399
- 1.0,
2400
- 1.0
2401
- ],
2402
- [
2403
- 3.0,
2404
- 1.0,
2405
- 1.0
2406
- ],
2407
- [
2408
- 3.0,
2409
- 1.0,
2410
- 1.0
2411
- ],
2412
- [
2413
- 3.0,
2414
- 1.0,
2415
- 1.0
2416
- ],
2417
- [
2418
- 3.0,
2419
- 1.0,
2420
- 1.0
2421
- ],
2422
- [
2423
- 3.0,
2424
- 1.0,
2425
- 1.0
2426
- ],
2427
- [
2428
- 3.0,
2429
- 1.0,
2430
- 1.0
2431
- ],
2432
- [
2433
- 3.0,
2434
- 1.0,
2435
- 1.0
2436
- ],
2437
- [
2438
- 3.0,
2439
- 1.0,
2440
- 1.0
2441
- ],
2442
- [
2443
- 3.0,
2444
- 1.0,
2445
- 1.0
2446
- ],
2447
- [
2448
- 3.0,
2449
- 1.0,
2450
- 1.0
2451
- ],
2452
- [
2453
- 3.0,
2454
- 1.0,
2455
- 1.0
2456
- ],
2457
- [
2458
- 3.0,
2459
- 1.0,
2460
- 1.0
2461
- ],
2462
- [
2463
- 3.0,
2464
- 1.0,
2465
- 1.0
2466
- ],
2467
- [
2468
- 3.0,
2469
- 1.0,
2470
- 1.0
2471
- ],
2472
- [
2473
- 3.0,
2474
- 1.0,
2475
- 1.0
2476
- ],
2477
- [
2478
- 3.0,
2479
- 1.0,
2480
- 1.0
2481
- ],
2482
- [
2483
- 3.0,
2484
- 1.0,
2485
- 1.0
2486
- ],
2487
- [
2488
- 3.0,
2489
- 1.0,
2490
- 1.0
2491
- ],
2492
- [
2493
- 3.0,
2494
- 1.0,
2495
- 1.0
2496
- ],
2497
- [
2498
- 3.0,
2499
- 1.0,
2500
- 1.0
2501
- ],
2502
- [
2503
- 3.0,
2504
- 1.0,
2505
- 1.0
2506
- ],
2507
- [
2508
- 3.0,
2509
- 1.0,
2510
- 1.0
2511
- ],
2512
- [
2513
- 3.0,
2514
- 1.0,
2515
- 1.0
2516
- ],
2517
- [
2518
- 3.0,
2519
- 1.0,
2520
- 1.0
2521
- ],
2522
- [
2523
- 3.0,
2524
- 1.0,
2525
- 1.0
2526
- ],
2527
- [
2528
- 3.0,
2529
- 1.0,
2530
- 1.0
2531
- ],
2532
- [
2533
- 3.0,
2534
- 1.0,
2535
- 1.0
2536
- ],
2537
- [
2538
- 3.0,
2539
- 1.0,
2540
- 1.0
2541
- ],
2542
- [
2543
- 3.0,
2544
- 1.0,
2545
- 1.0
2546
- ],
2547
- [
2548
- 3.0,
2549
- 1.0,
2550
- 1.0
2551
- ],
2552
- [
2553
- 3.0,
2554
- 1.0,
2555
- 1.0
2556
- ],
2557
- [
2558
- 3.0,
2559
- 1.0,
2560
- 1.0
2561
- ],
2562
- [
2563
- 3.0,
2564
- 1.0,
2565
- 1.0
2566
- ],
2567
- [
2568
- 3.0,
2569
- 1.0,
2570
- 1.0
2571
- ],
2572
- [
2573
- 3.0,
2574
- 1.0,
2575
- 1.0
2576
- ],
2577
- [
2578
- 3.0,
2579
- 1.0,
2580
- 1.0
2581
- ],
2582
- [
2583
- 3.0,
2584
- 1.0,
2585
- 1.0
2586
- ],
2587
- [
2588
- 3.0,
2589
- 1.0,
2590
- 1.0
2591
- ],
2592
- [
2593
- 3.0,
2594
- 1.0,
2595
- 1.0
2596
- ],
2597
- [
2598
- 3.0,
2599
- 1.0,
2600
- 1.0
2601
- ],
2602
- [
2603
- 3.0,
2604
- 1.0,
2605
- 1.0
2606
- ],
2607
- [
2608
- 3.0,
2609
- 1.0,
2610
- 1.0
2611
- ],
2612
- [
2613
- 3.0,
2614
- 1.0,
2615
- 1.0
2616
- ],
2617
- [
2618
- 3.0,
2619
- 1.0,
2620
- 1.0
2621
- ],
2622
- [
2623
- 3.0,
2624
- 1.0,
2625
- 1.0
2626
- ],
2627
- [
2628
- 3.0,
2629
- 1.0,
2630
- 1.0
2631
- ],
2632
- [
2633
- 3.0,
2634
- 1.0,
2635
- 1.0
2636
- ],
2637
- [
2638
- 3.0,
2639
- 1.0,
2640
- 1.0
2641
- ],
2642
- [
2643
- 3.0,
2644
- 1.0,
2645
- 1.0
2646
- ],
2647
- [
2648
- 3.0,
2649
- 1.0,
2650
- 1.0
2651
- ],
2652
- [
2653
- 3.0,
2654
- 1.0,
2655
- 1.0
2656
- ],
2657
- [
2658
- 3.0,
2659
- 1.0,
2660
- 1.0
2661
- ],
2662
- [
2663
- 3.0,
2664
- 1.0,
2665
- 1.0
2666
- ],
2667
- [
2668
- 3.0,
2669
- 1.0,
2670
- 1.0
2671
- ],
2672
- [
2673
- 3.0,
2674
- 1.0,
2675
- 1.0
2676
- ],
2677
- [
2678
- 3.0,
2679
- 1.0,
2680
- 1.0
2681
- ],
2682
- [
2683
- 3.0,
2684
- 1.0,
2685
- 1.0
2686
- ],
2687
- [
2688
- 3.0,
2689
- 1.0,
2690
- 1.0
2691
- ],
2692
- [
2693
- 3.0,
2694
- 1.0,
2695
- 1.0
2696
- ],
2697
- [
2698
- 3.0,
2699
- 1.0,
2700
- 1.0
2701
- ],
2702
- [
2703
- 3.0,
2704
- 1.0,
2705
- 1.0
2706
- ],
2707
- [
2708
- 3.0,
2709
- 1.0,
2710
- 1.0
2711
- ],
2712
- [
2713
- 3.0,
2714
- 1.0,
2715
- 1.0
2716
- ],
2717
- [
2718
- 3.0,
2719
- 1.0,
2720
- 1.0
2721
- ],
2722
- [
2723
- 3.0,
2724
- 1.0,
2725
- 1.0
2726
- ],
2727
- [
2728
- 3.0,
2729
- 1.0,
2730
- 1.0
2731
- ],
2732
- [
2733
- 3.0,
2734
- 1.0,
2735
- 1.0
2736
- ],
2737
- [
2738
- 3.0,
2739
- 1.0,
2740
- 1.0
2741
- ],
2742
- [
2743
- 3.0,
2744
- 1.0,
2745
- 1.0
2746
- ],
2747
- [
2748
- 3.0,
2749
- 1.0,
2750
- 1.0
2751
- ],
2752
- [
2753
- 3.0,
2754
- 1.0,
2755
- 1.0
2756
- ],
2757
- [
2758
- 3.0,
2759
- 1.0,
2760
- 1.0
2761
- ],
2762
- [
2763
- 3.0,
2764
- 1.0,
2765
- 1.0
2766
- ],
2767
- [
2768
- 3.0,
2769
- 1.0,
2770
- 1.0
2771
- ],
2772
- [
2773
- 3.0,
2774
- 1.0,
2775
- 1.0
2776
- ],
2777
- [
2778
- 3.0,
2779
- 1.0,
2780
- 1.0
2781
- ],
2782
- [
2783
- 3.0,
2784
- 1.0,
2785
- 1.0
2786
- ],
2787
- [
2788
- 3.0,
2789
- 1.0,
2790
- 1.0
2791
- ],
2792
- [
2793
- 3.0,
2794
- 1.0,
2795
- 1.0
2796
- ],
2797
- [
2798
- 3.0,
2799
- 1.0,
2800
- 1.0
2801
- ],
2802
- [
2803
- 3.0,
2804
- 1.0,
2805
- 1.0
2806
- ],
2807
- [
2808
- 3.0,
2809
- 1.0,
2810
- 1.0
2811
- ],
2812
- [
2813
- 3.0,
2814
- 1.0,
2815
- 1.0
2816
- ],
2817
- [
2818
- 3.0,
2819
- 1.0,
2820
- 1.0
2821
- ],
2822
- [
2823
- 3.0,
2824
- 1.0,
2825
- 1.0
2826
- ],
2827
- [
2828
- 3.0,
2829
- 1.0,
2830
- 1.0
2831
- ],
2832
- [
2833
- 3.0,
2834
- 1.0,
2835
- 1.0
2836
- ],
2837
- [
2838
- 3.0,
2839
- 1.0,
2840
- 1.0
2841
- ],
2842
- [
2843
- 3.0,
2844
- 1.0,
2845
- 1.0
2846
- ],
2847
- [
2848
- 3.0,
2849
- 1.0,
2850
- 1.0
2851
- ],
2852
- [
2853
- 3.0,
2854
- 1.0,
2855
- 1.0
2856
- ],
2857
- [
2858
- 3.0,
2859
- 1.0,
2860
- 1.0
2861
- ],
2862
- [
2863
- 3.0,
2864
- 1.0,
2865
- 1.0
2866
- ],
2867
- [
2868
- 3.0,
2869
- 1.0,
2870
- 1.0
2871
- ],
2872
- [
2873
- 3.0,
2874
- 1.0,
2875
- 1.0
2876
- ],
2877
- [
2878
- 3.0,
2879
- 1.0,
2880
- 1.0
2881
- ],
2882
- [
2883
- 3.0,
2884
- 1.0,
2885
- 1.0
2886
- ],
2887
- [
2888
- 3.0,
2889
- 1.0,
2890
- 1.0
2891
- ],
2892
- [
2893
- 3.0,
2894
- 1.0,
2895
- 1.0
2896
- ],
2897
- [
2898
- 3.0,
2899
- 1.0,
2900
- 1.0
2901
- ],
2902
- [
2903
- 3.0,
2904
- 1.0,
2905
- 1.0
2906
- ],
2907
- [
2908
- 3.0,
2909
- 1.0,
2910
- 1.0
2911
- ],
2912
- [
2913
- 3.0,
2914
- 1.0,
2915
- 1.0
2916
- ],
2917
- [
2918
- 3.0,
2919
- 1.0,
2920
- 1.0
2921
- ],
2922
- [
2923
- 3.0,
2924
- 1.0,
2925
- 1.0
2926
- ],
2927
- [
2928
- 3.0,
2929
- 1.0,
2930
- 1.0
2931
- ],
2932
- [
2933
- 3.0,
2934
- 1.0,
2935
- 1.0
2936
- ],
2937
- [
2938
- 3.0,
2939
- 1.0,
2940
- 1.0
2941
- ],
2942
- [
2943
- 3.0,
2944
- 1.0,
2945
- 1.0
2946
- ],
2947
- [
2948
- 3.0,
2949
- 1.0,
2950
- 1.0
2951
- ],
2952
- [
2953
- 3.0,
2954
- 1.0,
2955
- 1.0
2956
- ],
2957
- [
2958
- 3.0,
2959
- 1.0,
2960
- 1.0
2961
- ],
2962
- [
2963
- 3.0,
2964
- 1.0,
2965
- 1.0
2966
- ],
2967
- [
2968
- 3.0,
2969
- 1.0,
2970
- 1.0
2971
- ],
2972
- [
2973
- 3.0,
2974
- 1.0,
2975
- 1.0
2976
- ],
2977
- [
2978
- 3.0,
2979
- 1.0,
2980
- 1.0
2981
- ],
2982
- [
2983
- 3.0,
2984
- 1.0,
2985
- 1.0
2986
- ],
2987
- [
2988
- 3.0,
2989
- 1.0,
2990
- 1.0
2991
- ],
2992
- [
2993
- 3.0,
2994
- 1.0,
2995
- 1.0
2996
- ],
2997
- [
2998
- 3.0,
2999
- 1.0,
3000
- 1.0
3001
- ],
3002
- [
3003
- 3.0,
3004
- 1.0,
3005
- 1.0
3006
- ],
3007
- [
3008
- 3.0,
3009
- 1.0,
3010
- 1.0
3011
- ],
3012
- [
3013
- 3.0,
3014
- 1.0,
3015
- 1.0
3016
- ],
3017
- [
3018
- 3.0,
3019
- 1.0,
3020
- 1.0
3021
- ],
3022
- [
3023
- 3.0,
3024
- 1.0,
3025
- 1.0
3026
- ],
3027
- [
3028
- 3.0,
3029
- 1.0,
3030
- 1.0
3031
- ],
3032
- [
3033
- 3.0,
3034
- 1.0,
3035
- 1.0
3036
- ],
3037
- [
3038
- 3.0,
3039
- 1.0,
3040
- 1.0
3041
- ],
3042
- [
3043
- 3.0,
3044
- 1.0,
3045
- 1.0
3046
- ],
3047
- [
3048
- 3.0,
3049
- 1.0,
3050
- 1.0
3051
- ],
3052
- [
3053
- 3.0,
3054
- 1.0,
3055
- 1.0
3056
- ],
3057
- [
3058
- 3.0,
3059
- 1.0,
3060
- 1.0
3061
- ],
3062
- [
3063
- 3.0,
3064
- 1.0,
3065
- 1.0
3066
- ],
3067
- [
3068
- 3.0,
3069
- 1.0,
3070
- 1.0
3071
- ],
3072
- [
3073
- 3.0,
3074
- 1.0,
3075
- 1.0
3076
- ],
3077
- [
3078
- 3.0,
3079
- 1.0,
3080
- 1.0
3081
- ],
3082
- [
3083
- 3.0,
3084
- 1.0,
3085
- 1.0
3086
- ],
3087
- [
3088
- 3.0,
3089
- 1.0,
3090
- 1.0
3091
- ],
3092
- [
3093
- 3.0,
3094
- 1.0,
3095
- 1.0
3096
- ],
3097
- [
3098
- 3.0,
3099
- 1.0,
3100
- 1.0
3101
- ],
3102
- [
3103
- 3.0,
3104
- 1.0,
3105
- 1.0
3106
- ],
3107
- [
3108
- 3.0,
3109
- 1.0,
3110
- 1.0
3111
- ],
3112
- [
3113
- 3.0,
3114
- 1.0,
3115
- 1.0
3116
- ],
3117
- [
3118
- 3.0,
3119
- 1.0,
3120
- 1.0
3121
- ],
3122
- [
3123
- 3.0,
3124
- 1.0,
3125
- 1.0
3126
- ],
3127
- [
3128
- 3.0,
3129
- 1.0,
3130
- 1.0
3131
- ],
3132
- [
3133
- 3.0,
3134
- 1.0,
3135
- 1.0
3136
- ],
3137
- [
3138
- 3.0,
3139
- 1.0,
3140
- 1.0
3141
- ],
3142
- [
3143
- 3.0,
3144
- 1.0,
3145
- 1.0
3146
- ],
3147
- [
3148
- 3.0,
3149
- 1.0,
3150
- 1.0
3151
- ],
3152
- [
3153
- 3.0,
3154
- 1.0,
3155
- 1.0
3156
- ],
3157
- [
3158
- 3.0,
3159
- 1.0,
3160
- 1.0
3161
- ],
3162
- [
3163
- 3.0,
3164
- 1.0,
3165
- 1.0
3166
- ],
3167
- [
3168
- 3.0,
3169
- 1.0,
3170
- 1.0
3171
- ],
3172
- [
3173
- 3.0,
3174
- 1.0,
3175
- 1.0
3176
- ],
3177
- [
3178
- 3.0,
3179
- 1.0,
3180
- 1.0
3181
- ],
3182
- [
3183
- 3.0,
3184
- 1.0,
3185
- 1.0
3186
- ],
3187
- [
3188
- 3.0,
3189
- 1.0,
3190
- 1.0
3191
- ],
3192
- [
3193
- 3.0,
3194
- 1.0,
3195
- 1.0
3196
- ],
3197
- [
3198
- 3.0,
3199
- 1.0,
3200
- 1.0
3201
- ],
3202
- [
3203
- 3.0,
3204
- 1.0,
3205
- 1.0
3206
- ],
3207
- [
3208
- 3.0,
3209
- 1.0,
3210
- 1.0
3211
- ],
3212
- [
3213
- 3.0,
3214
- 1.0,
3215
- 1.0
3216
- ],
3217
- [
3218
- 3.0,
3219
- 1.0,
3220
- 1.0
3221
- ],
3222
- [
3223
- 3.0,
3224
- 1.0,
3225
- 1.0
3226
- ],
3227
- [
3228
- 3.0,
3229
- 1.0,
3230
- 1.0
3231
- ],
3232
- [
3233
- 3.0,
3234
- 1.0,
3235
- 1.0
3236
- ],
3237
- [
3238
- 3.0,
3239
- 1.0,
3240
- 1.0
3241
- ],
3242
- [
3243
- 3.0,
3244
- 1.0,
3245
- 1.0
3246
- ],
3247
- [
3248
- 3.0,
3249
- 1.0,
3250
- 1.0
3251
- ],
3252
- [
3253
- 3.0,
3254
- 1.0,
3255
- 1.0
3256
- ],
3257
- [
3258
- 3.0,
3259
- 1.0,
3260
- 1.0
3261
- ],
3262
- [
3263
- 3.0,
3264
- 1.0,
3265
- 1.0
3266
- ]
3267
- ]
3268
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_final.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ced0929a6ae2327f388ca4bb74dfca15acc111a68340b10442a2840895b5b5d8
3
- size 815577709
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json DELETED
@@ -1,58 +0,0 @@
1
- {
2
- "_best_ema": "0.047194228",
3
- "aim_run": "<Run#-1202935887677095771 name=7ec9a6e8af2847c58549736c repo=<Repo#-3891852473720853603 path=/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/runs_aim/.aim read_only=None>>",
4
- "batch_size": "2",
5
- "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': False}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
6
- "configuration_name": "3d_fullres",
7
- "cudnn_version": 90501,
8
- "current_epoch": "27",
9
- "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x15543c17dc50>",
10
- "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551f9793150>",
11
- "dataloader_train.num_processes": "12",
12
- "dataloader_train.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
13
- "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x15543c17f1d0>",
14
- "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551f97937d0>",
15
- "dataloader_val.num_processes": "6",
16
- "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
17
- "dataset_json": "{'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}",
18
- "debug_custom": "{'run_hash': 'de0b335742114ac0a04510ad'}",
19
- "decoder_type": "standard",
20
- "device": "cuda:0",
21
- "disable_checkpointing": "False",
22
- "enable_deep_supervision": "False",
23
- "fold": "0",
24
- "folder_with_segs_from_previous_stage": "None",
25
- "gpu_name": "NVIDIA H100",
26
- "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x15543db00690>",
27
- "hostname": "g030",
28
- "image_loss_weight": "0.5",
29
- "inference_allowed_mirroring_axes": "None",
30
- "initial_lr": "0.01",
31
- "is_cascaded": "False",
32
- "is_ddp": "False",
33
- "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x15543db00790>",
34
- "local_rank": "0",
35
- "log_file": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2025_8_14_18_25_01.txt",
36
- "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x15543e6adb50>",
37
- "loss": "SynPerceptionLoss(\n (seg_model): ResidualEncoderUNet(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (decoder): UNetDecoder(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (stages): ModuleList(\n (0): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (1): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (2): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (3): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (4): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (transpconvs): ModuleList(\n (0): ConvTranspose3d(320, 320, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n (1): ConvTranspose3d(320, 256, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (2): ConvTranspose3d(256, 128, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (3): ConvTranspose3d(128, 64, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (4): ConvTranspose3d(64, 32, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n )\n (seg_layers): ModuleList(\n (0): Conv3d(320, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (1): Conv3d(256, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (2): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (3): Conv3d(64, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (4): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n )\n )\n )\n (L1): L1Loss()\n (image_loss): myMaskedMSE(\n (mse): myMSE()\n )\n)",
38
- "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x15543c17d590>",
39
- "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}, 'configuration': '3d_fullres', 'fold': 0, 'dataset_json': {'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}, 'unpack_dataset': True, 'device': device(type='cuda')}",
40
- "network": "ResidualEncoderUNet",
41
- "num_epochs": "1000",
42
- "num_input_channels": "1",
43
- "num_iterations_per_epoch": "250",
44
- "num_val_iterations_per_epoch": "50",
45
- "optimizer": "SGD (\nParameter Group 0\n dampening: 0\n differentiable: False\n foreach: None\n fused: None\n initial_lr: 0.01\n lr: 0.009765692859724779\n maximize: False\n momentum: 0.99\n nesterov: True\n weight_decay: 3e-05\n)",
46
- "output_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0",
47
- "output_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres",
48
- "oversample_foreground_percent": "0.33",
49
- "perception_masked": "True",
50
- "plans_manager": "{'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}",
51
- "preprocessed_dataset_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetPlans_3d_fullres",
52
- "preprocessed_dataset_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
53
- "save_every": "50",
54
- "torch_version": "2.7.1+cu126",
55
- "unpack_dataset": "True",
56
- "was_initialized": "True",
57
- "weight_decay": "3e-05"
58
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png DELETED

Git LFS Details

  • SHA256: 5d87d6b783b54c578c4fb7db56c814d908815889146a7c6aad34673d1790dadc
  • Pointer size: 131 Bytes
  • Size of remote file: 491 kB
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_0/validation/summary.json DELETED
@@ -1,52 +0,0 @@
1
- {
2
- "mae": {
3
- "mean": 59.6831403703124,
4
- "std": 13.282268575250406,
5
- "max": 135.26334083952912,
6
- "min": 43.70561993673689,
7
- "25pc": 50.90544320106635,
8
- "50pc": 57.59766416973342,
9
- "75pc": 63.61140017268492,
10
- "count": 65
11
- },
12
- "psnr": {
13
- "mean": 30.41879186477032,
14
- "std": 1.7822310331910258,
15
- "max": 33.303479514662435,
16
- "min": 21.938598217631913,
17
- "25pc": 29.489967207335592,
18
- "50pc": 30.549653942101994,
19
- "75pc": 31.678537891293097,
20
- "count": 65
21
- },
22
- "ms_ssim": {
23
- "mean": 0.972537866604284,
24
- "std": 0.023400539772748585,
25
- "max": 0.9896922348656185,
26
- "min": 0.8256950304841117,
27
- "25pc": 0.972346153083238,
28
- "50pc": 0.9793511650351168,
29
- "75pc": 0.9834134511089297,
30
- "count": 65
31
- },
32
- "DICE": {
33
- "mean": 0.8757181415191063,
34
- "std": 0.04152858997642459,
35
- "max": 0.9331912994384766,
36
- "min": 0.7493325471878052,
37
- "25pc": 0.8536831736564636,
38
- "50pc": 0.885571300983429,
39
- "75pc": 0.907253623008728,
40
- "count": 65
41
- },
42
- "HD95": {
43
- "mean": 1.9038109889397254,
44
- "std": 0.8689346520839407,
45
- "max": 4.7832865715026855,
46
- "min": 1.1818181276321411,
47
- "25pc": 1.3928370475769043,
48
- "50pc": 1.54265558719635,
49
- "75pc": 2.0603861808776855,
50
- "count": 65
51
- }
52
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/checkpoint_final.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:5cfa7401d581fc33194e37d16a19c56de5d092c8a2e5c43f517e7815903fd24e
3
- size 815577901
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/debug.json DELETED
@@ -1,58 +0,0 @@
1
- {
2
- "_best_ema": "None",
3
- "aim_run": "<Run#868470907629722261 name=d1d441fac90f48bb98e95af7 repo=<Repo#6354568122047407417 path=/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/runs_aim/.aim read_only=None>>",
4
- "batch_size": "2",
5
- "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': False}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
6
- "configuration_name": "3d_fullres",
7
- "cudnn_version": 90501,
8
- "current_epoch": "0",
9
- "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x15543e72fdd0>",
10
- "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551e1737b50>",
11
- "dataloader_train.num_processes": "12",
12
- "dataloader_train.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
13
- "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1554407a0150>",
14
- "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551e1737f90>",
15
- "dataloader_val.num_processes": "6",
16
- "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
17
- "dataset_json": "{'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}",
18
- "debug_custom": "{'run_hash': 'd1d441fac90f48bb98e95af7'}",
19
- "decoder_type": "standard",
20
- "device": "cuda:0",
21
- "disable_checkpointing": "False",
22
- "enable_deep_supervision": "False",
23
- "fold": "1",
24
- "folder_with_segs_from_previous_stage": "None",
25
- "gpu_name": "NVIDIA H100",
26
- "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x15543e745210>",
27
- "hostname": "g078",
28
- "image_loss_weight": "0.5",
29
- "inference_allowed_mirroring_axes": "None",
30
- "initial_lr": "0.01",
31
- "is_cascaded": "False",
32
- "is_ddp": "False",
33
- "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x15543e693190>",
34
- "local_rank": "0",
35
- "log_file": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/training_log_2025_8_14_18_28_35.txt",
36
- "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x15543e7450d0>",
37
- "loss": "SynPerceptionLoss(\n (seg_model): ResidualEncoderUNet(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (decoder): UNetDecoder(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (stages): ModuleList(\n (0): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (1): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (2): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (3): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (4): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (transpconvs): ModuleList(\n (0): ConvTranspose3d(320, 320, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n (1): ConvTranspose3d(320, 256, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (2): ConvTranspose3d(256, 128, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (3): ConvTranspose3d(128, 64, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (4): ConvTranspose3d(64, 32, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n )\n (seg_layers): ModuleList(\n (0): Conv3d(320, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (1): Conv3d(256, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (2): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (3): Conv3d(64, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (4): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n )\n )\n )\n (L1): L1Loss()\n (image_loss): myMaskedMSE(\n (mse): myMSE()\n )\n)",
38
- "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x15543c1891d0>",
39
- "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}, 'configuration': '3d_fullres', 'fold': 1, 'dataset_json': {'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}, 'unpack_dataset': True, 'device': device(type='cuda')}",
40
- "network": "ResidualEncoderUNet",
41
- "num_epochs": "1000",
42
- "num_input_channels": "1",
43
- "num_iterations_per_epoch": "250",
44
- "num_val_iterations_per_epoch": "50",
45
- "optimizer": "SGD (\nParameter Group 0\n dampening: 0\n differentiable: False\n foreach: None\n fused: None\n initial_lr: 0.01\n lr: 0.01\n maximize: False\n momentum: 0.99\n nesterov: True\n weight_decay: 3e-05\n)",
46
- "output_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1",
47
- "output_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres",
48
- "oversample_foreground_percent": "0.33",
49
- "perception_masked": "True",
50
- "plans_manager": "{'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}",
51
- "preprocessed_dataset_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetPlans_3d_fullres",
52
- "preprocessed_dataset_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
53
- "save_every": "50",
54
- "torch_version": "2.7.1+cu126",
55
- "unpack_dataset": "True",
56
- "was_initialized": "True",
57
- "weight_decay": "3e-05"
58
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/progress.png DELETED

Git LFS Details

  • SHA256: 1d46d80bbd9214ebec78fe145b8bf6f1d2d21a8664f7e649938fd32df553c64b
  • Pointer size: 131 Bytes
  • Size of remote file: 482 kB
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_1/validation/summary.json DELETED
@@ -1,52 +0,0 @@
1
- {
2
- "mae": {
3
- "mean": 59.744778705390495,
4
- "std": 8.48563766360584,
5
- "max": 80.48682167216784,
6
- "min": 36.72710980567932,
7
- "25pc": 55.020186602101674,
8
- "50pc": 59.03544285077603,
9
- "75pc": 64.28191801866691,
10
- "count": 65
11
- },
12
- "psnr": {
13
- "mean": 30.29001651555625,
14
- "std": 1.2492612858694339,
15
- "max": 34.27661718381824,
16
- "min": 27.032412629993978,
17
- "25pc": 29.654636947017195,
18
- "50pc": 30.343156013061762,
19
- "75pc": 30.98505758413895,
20
- "count": 65
21
- },
22
- "ms_ssim": {
23
- "mean": 0.9760874668745824,
24
- "std": 0.011911759226168885,
25
- "max": 0.9945520598588762,
26
- "min": 0.9332606876848101,
27
- "25pc": 0.9748705939992588,
28
- "50pc": 0.9795418266521647,
29
- "75pc": 0.9826530409619525,
30
- "count": 65
31
- },
32
- "DICE": {
33
- "mean": 0.8751137064053462,
34
- "std": 0.05383460132488653,
35
- "max": 0.9602877497673035,
36
- "min": 0.7026435732841492,
37
- "25pc": 0.8616073131561279,
38
- "50pc": 0.8879862427711487,
39
- "75pc": 0.9115713834762573,
40
- "count": 65
41
- },
42
- "HD95": {
43
- "mean": 1.7658534985322218,
44
- "std": 0.6488990819765412,
45
- "max": 4.268579959869385,
46
- "min": 1.0,
47
- "25pc": 1.3333333730697632,
48
- "50pc": 1.5522551536560059,
49
- "75pc": 1.9791498184204102,
50
- "count": 65
51
- }
52
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/checkpoint_final.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:9726b055997662a73b2a6f3b889f84203f93baf6b77c9630bb701aea227b738b
3
- size 815577965
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/debug.json DELETED
@@ -1,58 +0,0 @@
1
- {
2
- "_best_ema": "None",
3
- "aim_run": "<Run#4755309242488009678 name=cc34edb573374b4f87d07317 repo=<Repo#1363373405569143032 path=/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/runs_aim/.aim read_only=None>>",
4
- "batch_size": "2",
5
- "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': False}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
6
- "configuration_name": "3d_fullres",
7
- "cudnn_version": 90501,
8
- "current_epoch": "0",
9
- "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x155230558090>",
10
- "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x155215f43e10>",
11
- "dataloader_train.num_processes": "12",
12
- "dataloader_train.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
13
- "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1552305826d0>",
14
- "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x15543dc19d90>",
15
- "dataloader_val.num_processes": "6",
16
- "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
17
- "dataset_json": "{'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}",
18
- "debug_custom": "{'run_hash': 'cc34edb573374b4f87d07317'}",
19
- "decoder_type": "standard",
20
- "device": "cuda:0",
21
- "disable_checkpointing": "False",
22
- "enable_deep_supervision": "False",
23
- "fold": "2",
24
- "folder_with_segs_from_previous_stage": "None",
25
- "gpu_name": "NVIDIA H100",
26
- "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x15543db00310>",
27
- "hostname": "g079",
28
- "image_loss_weight": "0.5",
29
- "inference_allowed_mirroring_axes": "None",
30
- "initial_lr": "0.01",
31
- "is_cascaded": "False",
32
- "is_ddp": "False",
33
- "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x15543db00250>",
34
- "local_rank": "0",
35
- "log_file": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/training_log_2025_8_14_18_28_34.txt",
36
- "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x15543e70db50>",
37
- "loss": "SynPerceptionLoss(\n (seg_model): ResidualEncoderUNet(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (decoder): UNetDecoder(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (stages): ModuleList(\n (0): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (1): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (2): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (3): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (4): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (transpconvs): ModuleList(\n (0): ConvTranspose3d(320, 320, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n (1): ConvTranspose3d(320, 256, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (2): ConvTranspose3d(256, 128, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (3): ConvTranspose3d(128, 64, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (4): ConvTranspose3d(64, 32, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n )\n (seg_layers): ModuleList(\n (0): Conv3d(320, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (1): Conv3d(256, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (2): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (3): Conv3d(64, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (4): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n )\n )\n )\n (L1): L1Loss()\n (image_loss): myMaskedMSE(\n (mse): myMSE()\n )\n)",
38
- "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x15543c17ae10>",
39
- "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}, 'configuration': '3d_fullres', 'fold': 2, 'dataset_json': {'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}, 'unpack_dataset': True, 'device': device(type='cuda')}",
40
- "network": "ResidualEncoderUNet",
41
- "num_epochs": "1000",
42
- "num_input_channels": "1",
43
- "num_iterations_per_epoch": "250",
44
- "num_val_iterations_per_epoch": "50",
45
- "optimizer": "SGD (\nParameter Group 0\n dampening: 0\n differentiable: False\n foreach: None\n fused: None\n initial_lr: 0.01\n lr: 0.01\n maximize: False\n momentum: 0.99\n nesterov: True\n weight_decay: 3e-05\n)",
46
- "output_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2",
47
- "output_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres",
48
- "oversample_foreground_percent": "0.33",
49
- "perception_masked": "True",
50
- "plans_manager": "{'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}",
51
- "preprocessed_dataset_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetPlans_3d_fullres",
52
- "preprocessed_dataset_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
53
- "save_every": "50",
54
- "torch_version": "2.7.1+cu126",
55
- "unpack_dataset": "True",
56
- "was_initialized": "True",
57
- "weight_decay": "3e-05"
58
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/progress.png DELETED

Git LFS Details

  • SHA256: 9cddf04336c342eff9d6a4e2e940b7f9d12746999c5b76bcb62d87eb7af76266
  • Pointer size: 131 Bytes
  • Size of remote file: 460 kB
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_2/validation/summary.json DELETED
@@ -1,52 +0,0 @@
1
- {
2
- "mae": {
3
- "mean": 59.439250763091884,
4
- "std": 9.144718015626115,
5
- "max": 78.70325904506652,
6
- "min": 42.80315891812898,
7
- "25pc": 52.068854137111984,
8
- "50pc": 58.805207557500694,
9
- "75pc": 64.72047789452675,
10
- "count": 65
11
- },
12
- "psnr": {
13
- "mean": 30.377276184144822,
14
- "std": 1.4223429950590885,
15
- "max": 32.8297584904791,
16
- "min": 27.075648135650216,
17
- "25pc": 29.40093059045853,
18
- "50pc": 30.543686856511627,
19
- "75pc": 31.403316426777565,
20
- "count": 65
21
- },
22
- "ms_ssim": {
23
- "mean": 0.9755103491560523,
24
- "std": 0.014031378198948008,
25
- "max": 0.9907865003729753,
26
- "min": 0.8955521856836826,
27
- "25pc": 0.9706492203118354,
28
- "50pc": 0.9792637813534784,
29
- "75pc": 0.983024084292943,
30
- "count": 65
31
- },
32
- "DICE": {
33
- "mean": 0.883325562110314,
34
- "std": 0.03812835864863602,
35
- "max": 0.9535425305366516,
36
- "min": 0.7242317199707031,
37
- "25pc": 0.8683831691741943,
38
- "50pc": 0.8890585899353027,
39
- "75pc": 0.9051895141601562,
40
- "count": 65
41
- },
42
- "HD95": {
43
- "mean": 1.813704824447632,
44
- "std": 1.0017171191332908,
45
- "max": 8.57068157196045,
46
- "min": 1.0,
47
- "25pc": 1.3902597427368164,
48
- "50pc": 1.619842290878296,
49
- "75pc": 1.8608479499816895,
50
- "count": 65
51
- }
52
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/checkpoint_final.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:7d90c4e21800bf53a7751295a84ab9b55a93aa07fac3120688f8ebbb48aed2c4
3
- size 815577773
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/debug.json DELETED
@@ -1,58 +0,0 @@
1
- {
2
- "_best_ema": "None",
3
- "aim_run": "<Run#5302827298361039645 name=0fe427721e024cb69e1cc584 repo=<Repo#-1279860204956534237 path=/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/runs_aim/.aim read_only=None>>",
4
- "batch_size": "2",
5
- "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': False}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
6
- "configuration_name": "3d_fullres",
7
- "cudnn_version": 90501,
8
- "current_epoch": "0",
9
- "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1551e5731fd0>",
10
- "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551e5731710>",
11
- "dataloader_train.num_processes": "12",
12
- "dataloader_train.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
13
- "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1551e5731750>",
14
- "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551e5730250>",
15
- "dataloader_val.num_processes": "6",
16
- "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
17
- "dataset_json": "{'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}",
18
- "debug_custom": "{'run_hash': '0fe427721e024cb69e1cc584'}",
19
- "decoder_type": "standard",
20
- "device": "cuda:0",
21
- "disable_checkpointing": "False",
22
- "enable_deep_supervision": "False",
23
- "fold": "3",
24
- "folder_with_segs_from_previous_stage": "None",
25
- "gpu_name": "NVIDIA H100",
26
- "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x15543daebf10>",
27
- "hostname": "g033",
28
- "image_loss_weight": "0.5",
29
- "inference_allowed_mirroring_axes": "None",
30
- "initial_lr": "0.01",
31
- "is_cascaded": "False",
32
- "is_ddp": "False",
33
- "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x15543daebe50>",
34
- "local_rank": "0",
35
- "log_file": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/training_log_2025_8_14_18_28_35.txt",
36
- "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x15543e71e490>",
37
- "loss": "SynPerceptionLoss(\n (seg_model): ResidualEncoderUNet(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (decoder): UNetDecoder(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (stages): ModuleList(\n (0): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (1): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (2): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (3): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (4): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (transpconvs): ModuleList(\n (0): ConvTranspose3d(320, 320, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n (1): ConvTranspose3d(320, 256, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (2): ConvTranspose3d(256, 128, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (3): ConvTranspose3d(128, 64, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (4): ConvTranspose3d(64, 32, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n )\n (seg_layers): ModuleList(\n (0): Conv3d(320, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (1): Conv3d(256, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (2): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (3): Conv3d(64, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (4): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n )\n )\n )\n (L1): L1Loss()\n (image_loss): myMaskedMSE(\n (mse): myMSE()\n )\n)",
38
- "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x15543c178990>",
39
- "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}, 'configuration': '3d_fullres', 'fold': 3, 'dataset_json': {'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}, 'unpack_dataset': True, 'device': device(type='cuda')}",
40
- "network": "ResidualEncoderUNet",
41
- "num_epochs": "1000",
42
- "num_input_channels": "1",
43
- "num_iterations_per_epoch": "250",
44
- "num_val_iterations_per_epoch": "50",
45
- "optimizer": "SGD (\nParameter Group 0\n dampening: 0\n differentiable: False\n foreach: None\n fused: None\n initial_lr: 0.01\n lr: 0.01\n maximize: False\n momentum: 0.99\n nesterov: True\n weight_decay: 3e-05\n)",
46
- "output_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3",
47
- "output_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres",
48
- "oversample_foreground_percent": "0.33",
49
- "perception_masked": "True",
50
- "plans_manager": "{'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}",
51
- "preprocessed_dataset_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetPlans_3d_fullres",
52
- "preprocessed_dataset_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
53
- "save_every": "50",
54
- "torch_version": "2.7.1+cu126",
55
- "unpack_dataset": "True",
56
- "was_initialized": "True",
57
- "weight_decay": "3e-05"
58
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/progress.png DELETED

Git LFS Details

  • SHA256: bc1627df80744f0707aab4875857ab67c7a0537d62d4ecde3f8895b0e6f8dc10
  • Pointer size: 131 Bytes
  • Size of remote file: 466 kB
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_3/validation/summary.json DELETED
@@ -1,52 +0,0 @@
1
- {
2
- "mae": {
3
- "mean": 59.47002310455084,
4
- "std": 10.81101956977336,
5
- "max": 94.50828806284945,
6
- "min": 36.317991411599486,
7
- "25pc": 52.56278780551801,
8
- "50pc": 59.594791153916205,
9
- "75pc": 66.75560383832976,
10
- "count": 65
11
- },
12
- "psnr": {
13
- "mean": 30.436606174983975,
14
- "std": 1.6599915209936646,
15
- "max": 33.89549634476005,
16
- "min": 25.09274031843838,
17
- "25pc": 29.529631611563087,
18
- "50pc": 30.392155700379533,
19
- "75pc": 31.561705183768176,
20
- "count": 65
21
- },
22
- "ms_ssim": {
23
- "mean": 0.9760188647757344,
24
- "std": 0.01237160436690702,
25
- "max": 0.9932310644312488,
26
- "min": 0.9152976435953994,
27
- "25pc": 0.970455666718718,
28
- "50pc": 0.9783151481112313,
29
- "75pc": 0.9842064148424343,
30
- "count": 65
31
- },
32
- "DICE": {
33
- "mean": 0.8854188469740061,
34
- "std": 0.04068928940244239,
35
- "max": 0.9550503492355347,
36
- "min": 0.7674918174743652,
37
- "25pc": 0.8717526197433472,
38
- "50pc": 0.888530433177948,
39
- "75pc": 0.9147038459777832,
40
- "count": 65
41
- },
42
- "HD95": {
43
- "mean": 1.7257186284432044,
44
- "std": 0.570431841226223,
45
- "max": 3.76181697845459,
46
- "min": 1.0,
47
- "25pc": 1.2859621047973633,
48
- "50pc": 1.5108829736709595,
49
- "75pc": 1.9506815671920776,
50
- "count": 65
51
- }
52
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/checkpoint_final.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:31734306539ca206a8a953bf04eb4cd2e1fe553606db7a2b26f40399983ae88e
3
- size 815577837
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/debug.json DELETED
@@ -1,58 +0,0 @@
1
- {
2
- "_best_ema": "None",
3
- "aim_run": "<Run#8165541130948727181 name=f0e89b2883824d1e97ac3301 repo=<Repo#3533591805119497480 path=/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/runs_aim/.aim read_only=None>>",
4
- "batch_size": "2",
5
- "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': False}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
6
- "configuration_name": "3d_fullres",
7
- "cudnn_version": 90501,
8
- "current_epoch": "0",
9
- "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1551e5735110>",
10
- "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x1551e572b590>",
11
- "dataloader_train.num_processes": "12",
12
- "dataloader_train.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
13
- "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x1551e5734bd0>",
14
- "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D_MRCT_mask object at 0x15543e70da50>",
15
- "dataloader_val.num_processes": "6",
16
- "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), NumpyToTensor( keys = ['data', 'target', 'mask'], cast_to = 'float' )] )",
17
- "dataset_json": "{'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}",
18
- "debug_custom": "{'run_hash': 'f0e89b2883824d1e97ac3301'}",
19
- "decoder_type": "standard",
20
- "device": "cuda:0",
21
- "disable_checkpointing": "False",
22
- "enable_deep_supervision": "False",
23
- "fold": "4",
24
- "folder_with_segs_from_previous_stage": "None",
25
- "gpu_name": "NVIDIA H100",
26
- "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x15543db00210>",
27
- "hostname": "g029",
28
- "image_loss_weight": "0.5",
29
- "inference_allowed_mirroring_axes": "None",
30
- "initial_lr": "0.01",
31
- "is_cascaded": "False",
32
- "is_ddp": "False",
33
- "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x15543db00150>",
34
- "local_rank": "0",
35
- "log_file": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/training_log_2025_8_14_18_28_34.txt",
36
- "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x15543e745590>",
37
- "loss": "SynPerceptionLoss(\n (seg_model): ResidualEncoderUNet(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (decoder): UNetDecoder(\n (encoder): ResidualEncoder(\n (stem): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(1, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (stages): Sequential(\n (0): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (1): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (2): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (3): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (4): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[2, 2, 2], stride=[2, 2, 2], padding=0)\n (1): ConvDropoutNormReLU(\n (conv): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(256, 320, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n (5): StackedResidualBlocks(\n (blocks): Sequential(\n (0): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 2, 2), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n (skip): Sequential(\n (0): AvgPool3d(kernel_size=[1, 2, 2], stride=[1, 2, 2], padding=0)\n )\n )\n (1): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (2): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (3): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (4): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n (5): BasicBlockD(\n (conv1): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n (conv2): ConvDropoutNormReLU(\n (conv): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (all_modules): Sequential(\n (0): Conv3d(320, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n )\n )\n (nonlin2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (stages): ModuleList(\n (0): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(640, 320, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (1): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(512, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (2): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(256, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (3): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (norm): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(128, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))\n (1): InstanceNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n (4): StackedConvBlocks(\n (convs): Sequential(\n (0): ConvDropoutNormReLU(\n (conv): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (norm): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (nonlin): LeakyReLU(negative_slope=0.01, inplace=True)\n (all_modules): Sequential(\n (0): Conv3d(64, 32, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1))\n (1): InstanceNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=False)\n (2): LeakyReLU(negative_slope=0.01, inplace=True)\n )\n )\n )\n )\n )\n (transpconvs): ModuleList(\n (0): ConvTranspose3d(320, 320, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n (1): ConvTranspose3d(320, 256, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (2): ConvTranspose3d(256, 128, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (3): ConvTranspose3d(128, 64, kernel_size=(2, 2, 2), stride=(2, 2, 2))\n (4): ConvTranspose3d(64, 32, kernel_size=(1, 2, 2), stride=(1, 2, 2))\n )\n (seg_layers): ModuleList(\n (0): Conv3d(320, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (1): Conv3d(256, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (2): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (3): Conv3d(64, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n (4): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))\n )\n )\n )\n (L1): L1Loss()\n (image_loss): myMaskedMSE(\n (mse): myMSE()\n )\n)",
38
- "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x15543e745950>",
39
- "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}, 'configuration': '3d_fullres', 'fold': 4, 'dataset_json': {'labels': {'label_001': '1', 'background': 0}, 'channel_names': {'0': 'CT_zscore_synthrad'}, 'numTraining': 325, 'file_ending': '.mha'}, 'unpack_dataset': True, 'device': device(type='cuda')}",
40
- "network": "ResidualEncoderUNet",
41
- "num_epochs": "1000",
42
- "num_input_channels": "1",
43
- "num_iterations_per_epoch": "250",
44
- "num_val_iterations_per_epoch": "50",
45
- "optimizer": "SGD (\nParameter Group 0\n dampening: 0\n differentiable: False\n foreach: None\n fused: None\n initial_lr: 0.01\n lr: 0.01\n maximize: False\n momentum: 0.99\n nesterov: True\n weight_decay: 3e-05\n)",
46
- "output_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4",
47
- "output_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/results/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres",
48
- "oversample_foreground_percent": "0.33",
49
- "perception_masked": "True",
50
- "plans_manager": "{'dataset_name': 'Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [3.0, 1.0, 1.0], 'original_median_shape_after_transp': [87, 308, 309], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 30, 'patch_size': [320, 320], 'median_image_size_in_voxels': [308.0, 309.0], 'spacing': [1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [48, 192, 192], 'median_image_size_in_voxels': [87.0, 308.0, 309.0], 'spacing': [3.0, 1.0, 1.0], 'normalization_schemes': ['CTNormalizationClippingSynthrad2025'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 3071.0, 'mean': -215.1082763671875, 'median': -133.0, 'min': -1024.0, 'percentile_00_5': -1024.0, 'percentile_99_5': 1345.0, 'std': 459.63665771484375}}}",
51
- "preprocessed_dataset_folder": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both/nnUNetPlans_3d_fullres",
52
- "preprocessed_dataset_folder_base": "/datasets/work/hb-synthrad2023/work/synthrad2025/bw_workplace/data/nnunet_struct/preprocessed/Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
53
- "save_every": "50",
54
- "torch_version": "2.7.1+cu126",
55
- "unpack_dataset": "True",
56
- "was_initialized": "True",
57
- "weight_decay": "3e-05"
58
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/progress.png DELETED

Git LFS Details

  • SHA256: dd4a9a0e599c30e4bd7a19a8413169a5970e0733defeb2729aed2ae20c58ad21
  • Pointer size: 131 Bytes
  • Size of remote file: 469 kB
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/fold_4/validation/summary.json DELETED
@@ -1,52 +0,0 @@
1
- {
2
- "mae": {
3
- "mean": 59.81550939299808,
4
- "std": 9.03344137096807,
5
- "max": 86.55407411236509,
6
- "min": 41.31046503426345,
7
- "25pc": 53.78254540290808,
8
- "50pc": 58.900544413718634,
9
- "75pc": 64.43899849257839,
10
- "count": 65
11
- },
12
- "psnr": {
13
- "mean": 30.27903952027891,
14
- "std": 1.4029605061057437,
15
- "max": 33.444057935542446,
16
- "min": 26.74722877488563,
17
- "25pc": 29.606531570444826,
18
- "50pc": 30.468317698487457,
19
- "75pc": 31.135384300124286,
20
- "count": 65
21
- },
22
- "ms_ssim": {
23
- "mean": 0.9748701394422348,
24
- "std": 0.01868224022307785,
25
- "max": 0.9919857072378632,
26
- "min": 0.8477223848270846,
27
- "25pc": 0.9713811703376288,
28
- "50pc": 0.9800244903527164,
29
- "75pc": 0.9839494228128256,
30
- "count": 65
31
- },
32
- "DICE": {
33
- "mean": 0.8786676654448876,
34
- "std": 0.04085292480312972,
35
- "max": 0.9457669854164124,
36
- "min": 0.7385271787643433,
37
- "25pc": 0.857272207736969,
38
- "50pc": 0.8850048184394836,
39
- "75pc": 0.9062296152114868,
40
- "count": 65
41
- },
42
- "HD95": {
43
- "mean": 1.8806897511849037,
44
- "std": 0.7390234507419097,
45
- "max": 4.64456844329834,
46
- "min": 1.0,
47
- "25pc": 1.3829010725021362,
48
- "50pc": 1.6428419351577759,
49
- "75pc": 1.9808140993118286,
50
- "count": 65
51
- }
52
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dataset542/nnUNetTrainerMRCT_loss_masked_perception_masked__nnUNetResEncUNetLPlans__3d_fullres/plans.json DELETED
@@ -1,356 +0,0 @@
1
- {
2
- "dataset_name": "Dataset542_synthrad2025_task2_CBCT_HN_pre_v2r_stitched_masked_both",
3
- "plans_name": "nnUNetResEncUNetLPlans",
4
- "original_median_spacing_after_transp": [
5
- 3.0,
6
- 1.0,
7
- 1.0
8
- ],
9
- "original_median_shape_after_transp": [
10
- 87,
11
- 308,
12
- 309
13
- ],
14
- "image_reader_writer": "SimpleITKIO",
15
- "transpose_forward": [
16
- 0,
17
- 1,
18
- 2
19
- ],
20
- "transpose_backward": [
21
- 0,
22
- 1,
23
- 2
24
- ],
25
- "configurations": {
26
- "2d": {
27
- "data_identifier": "nnUNetPlans_2d",
28
- "preprocessor_name": "DefaultPreprocessor",
29
- "batch_size": 30,
30
- "patch_size": [
31
- 320,
32
- 320
33
- ],
34
- "median_image_size_in_voxels": [
35
- 308.0,
36
- 309.0
37
- ],
38
- "spacing": [
39
- 1.0,
40
- 1.0
41
- ],
42
- "normalization_schemes": [
43
- "CTNormalizationClippingSynthrad2025"
44
- ],
45
- "use_mask_for_norm": [
46
- false
47
- ],
48
- "resampling_fn_data": "resample_data_or_seg_to_shape",
49
- "resampling_fn_seg": "resample_data_or_seg_to_shape",
50
- "resampling_fn_data_kwargs": {
51
- "is_seg": false,
52
- "order": 3,
53
- "order_z": 0,
54
- "force_separate_z": null
55
- },
56
- "resampling_fn_seg_kwargs": {
57
- "is_seg": true,
58
- "order": 1,
59
- "order_z": 0,
60
- "force_separate_z": null
61
- },
62
- "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
63
- "resampling_fn_probabilities_kwargs": {
64
- "is_seg": false,
65
- "order": 1,
66
- "order_z": 0,
67
- "force_separate_z": null
68
- },
69
- "architecture": {
70
- "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
71
- "arch_kwargs": {
72
- "n_stages": 7,
73
- "features_per_stage": [
74
- 32,
75
- 64,
76
- 128,
77
- 256,
78
- 512,
79
- 512,
80
- 512
81
- ],
82
- "conv_op": "torch.nn.modules.conv.Conv2d",
83
- "kernel_sizes": [
84
- [
85
- 3,
86
- 3
87
- ],
88
- [
89
- 3,
90
- 3
91
- ],
92
- [
93
- 3,
94
- 3
95
- ],
96
- [
97
- 3,
98
- 3
99
- ],
100
- [
101
- 3,
102
- 3
103
- ],
104
- [
105
- 3,
106
- 3
107
- ],
108
- [
109
- 3,
110
- 3
111
- ]
112
- ],
113
- "strides": [
114
- [
115
- 1,
116
- 1
117
- ],
118
- [
119
- 2,
120
- 2
121
- ],
122
- [
123
- 2,
124
- 2
125
- ],
126
- [
127
- 2,
128
- 2
129
- ],
130
- [
131
- 2,
132
- 2
133
- ],
134
- [
135
- 2,
136
- 2
137
- ],
138
- [
139
- 2,
140
- 2
141
- ]
142
- ],
143
- "n_blocks_per_stage": [
144
- 1,
145
- 3,
146
- 4,
147
- 6,
148
- 6,
149
- 6,
150
- 6
151
- ],
152
- "n_conv_per_stage_decoder": [
153
- 1,
154
- 1,
155
- 1,
156
- 1,
157
- 1,
158
- 1
159
- ],
160
- "conv_bias": true,
161
- "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
162
- "norm_op_kwargs": {
163
- "eps": 1e-05,
164
- "affine": true
165
- },
166
- "dropout_op": null,
167
- "dropout_op_kwargs": null,
168
- "nonlin": "torch.nn.LeakyReLU",
169
- "nonlin_kwargs": {
170
- "inplace": true
171
- }
172
- },
173
- "_kw_requires_import": [
174
- "conv_op",
175
- "norm_op",
176
- "dropout_op",
177
- "nonlin"
178
- ]
179
- },
180
- "batch_dice": true
181
- },
182
- "3d_fullres": {
183
- "data_identifier": "nnUNetPlans_3d_fullres",
184
- "preprocessor_name": "DefaultPreprocessor",
185
- "batch_size": 2,
186
- "patch_size": [
187
- 48,
188
- 192,
189
- 192
190
- ],
191
- "median_image_size_in_voxels": [
192
- 87.0,
193
- 308.0,
194
- 309.0
195
- ],
196
- "spacing": [
197
- 3.0,
198
- 1.0,
199
- 1.0
200
- ],
201
- "normalization_schemes": [
202
- "CTNormalizationClippingSynthrad2025"
203
- ],
204
- "use_mask_for_norm": [
205
- false
206
- ],
207
- "resampling_fn_data": "resample_data_or_seg_to_shape",
208
- "resampling_fn_seg": "resample_data_or_seg_to_shape",
209
- "resampling_fn_data_kwargs": {
210
- "is_seg": false,
211
- "order": 3,
212
- "order_z": 0,
213
- "force_separate_z": null
214
- },
215
- "resampling_fn_seg_kwargs": {
216
- "is_seg": true,
217
- "order": 1,
218
- "order_z": 0,
219
- "force_separate_z": null
220
- },
221
- "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
222
- "resampling_fn_probabilities_kwargs": {
223
- "is_seg": false,
224
- "order": 1,
225
- "order_z": 0,
226
- "force_separate_z": null
227
- },
228
- "architecture": {
229
- "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
230
- "arch_kwargs": {
231
- "n_stages": 6,
232
- "features_per_stage": [
233
- 32,
234
- 64,
235
- 128,
236
- 256,
237
- 320,
238
- 320
239
- ],
240
- "conv_op": "torch.nn.modules.conv.Conv3d",
241
- "kernel_sizes": [
242
- [
243
- 1,
244
- 3,
245
- 3
246
- ],
247
- [
248
- 3,
249
- 3,
250
- 3
251
- ],
252
- [
253
- 3,
254
- 3,
255
- 3
256
- ],
257
- [
258
- 3,
259
- 3,
260
- 3
261
- ],
262
- [
263
- 3,
264
- 3,
265
- 3
266
- ],
267
- [
268
- 3,
269
- 3,
270
- 3
271
- ]
272
- ],
273
- "strides": [
274
- [
275
- 1,
276
- 1,
277
- 1
278
- ],
279
- [
280
- 1,
281
- 2,
282
- 2
283
- ],
284
- [
285
- 2,
286
- 2,
287
- 2
288
- ],
289
- [
290
- 2,
291
- 2,
292
- 2
293
- ],
294
- [
295
- 2,
296
- 2,
297
- 2
298
- ],
299
- [
300
- 1,
301
- 2,
302
- 2
303
- ]
304
- ],
305
- "n_blocks_per_stage": [
306
- 1,
307
- 3,
308
- 4,
309
- 6,
310
- 6,
311
- 6
312
- ],
313
- "n_conv_per_stage_decoder": [
314
- 1,
315
- 1,
316
- 1,
317
- 1,
318
- 1
319
- ],
320
- "conv_bias": true,
321
- "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
322
- "norm_op_kwargs": {
323
- "eps": 1e-05,
324
- "affine": true
325
- },
326
- "dropout_op": null,
327
- "dropout_op_kwargs": null,
328
- "nonlin": "torch.nn.LeakyReLU",
329
- "nonlin_kwargs": {
330
- "inplace": true
331
- }
332
- },
333
- "_kw_requires_import": [
334
- "conv_op",
335
- "norm_op",
336
- "dropout_op",
337
- "nonlin"
338
- ]
339
- },
340
- "batch_dice": false
341
- }
342
- },
343
- "experiment_planner_used": "nnUNetPlannerResEncL",
344
- "label_manager": "LabelManager",
345
- "foreground_intensity_properties_per_channel": {
346
- "0": {
347
- "max": 3071.0,
348
- "mean": -215.1082763671875,
349
- "median": -133.0,
350
- "min": -1024.0,
351
- "percentile_00_5": -1024.0,
352
- "percentile_99_5": 1345.0,
353
- "std": 459.63665771484375
354
- }
355
- }
356
- }