-
Notifications
You must be signed in to change notification settings - Fork 1
/
data_access.py
2119 lines (1863 loc) · 86.9 KB
/
data_access.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# ================================================================================================ #
# Module for keeping track of processed data that is ready for analysis. Some classes for organizing
# data are included.
#
# Author: Eddie Lee edl56@cornell.edu
# ================================================================================================ #
from .utils import *
import os
import pickle as pickle
from warnings import warn
def subject_settings_v3(index,return_list=True):
settings = [{'person':'Zimu3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Darshna3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Richard3',
'modelhandedness':['Right','Left','Right','Left'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Rachel3',
'modelhandedness':['Right','Left','Right','Left'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Priyanka3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Emily3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Sam3',
'modelhandedness':['Right','Left','Right','Left'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Najila3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Kemper3',
'modelhandedness':['Left','Right','Left','Right'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']},
{'person':'Lauren3',
'modelhandedness':['Right','Left','Right','Left'],
'rotation':[0,0,0,0],
'trials':['avatar0','avatar1','hand0','hand1']}
][index]
dr = '../data/UE4_Experiments/%s'%settings['person']
if return_list:
output = [settings[k] for k in ['person','modelhandedness','rotation']]
output.append(dr)
return output
return settings,dr
def subject_settings_v3_1(index,return_list=True):
"""
Subject info for experiment v3.1.
Parameters
----------
index : int
return_list : bool,True
Returns
-------
settings : dict
dr : str
"""
settings = [{'person':'Subject01_3_1',
'modelhandedness':['Left'],
'rotation':[0],
'trials':['avatar']},
{'person':'Subject02_3_1',
'modelhandedness':['Left'],
'rotation':[0],
'trials':['avatar']},
{'person':'Subject03_3_1',
'modelhandedness':['Left'],
'rotation':[0],
'trials':['avatar']},
{'person':'Subject04_3_1',
'modelhandedness':['Right'],
'rotation':[0],
'trials':['avatar']},
{'person':'Subject05_3_1',
'modelhandedness':['Right'],
'rotation':[0],
'trials':['avatar']}
][index]
dr = '../data/UE4_Experiments/%s'%settings['person']
if return_list:
output = [settings[k] for k in ['person','modelhandedness','rotation']]
output.append(dr)
return output
return settings,dr
def subject_settings_v3_2(index,return_list=True):
"""
Subject info for experiment v3.2.
Parameters
----------
index : int
return_list : bool,True
Returns
-------
settings : dict
dr : str
"""
settings = [{'person':'Subject01_3_2',
'rotation':[0],
'trials':['avatar']},
{'person':'Subject02_3_2',
'rotation':[0],
'trials':['avatar']}
][index]
dr = '../data/UE4_Experiments/%s'%settings['person']
if return_list:
output = [settings[k] for k in ['person','rotation']]
output.append(dr)
return output
return settings,dr
def subject_settings_v3_3(index,hand,return_list=True):
"""
Subject info for experiment v3.3. Twoples refer to the left and right subject hands.
2018-01-15
Parameters
----------
index : int
hand : str
Subject hand used.
return_list : bool,True
Returns
-------
settings : dict
dr : str
"""
settings = [{'person':'Subject01_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject02_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject03_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject04_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject05_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject06_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject07_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject08_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject09_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject10_3_3',
'trials':['avatar'],
'reverse':[False,True],
'usable':[False,True]},
{'person':'Subject11_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,False]},
{'person':'Subject12_3_3',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
][index]
dr = '../data/UE4_Experiments/%s/%s'%(settings['person'],hand)
try:
rotAngle = pickle.load(open('%s/%s'%(dr,'gpr.p'),'rb'),encoding='latin1')['rotAngle']
except IOError:
rotAngle=np.nan
reverse=settings['reverse'][0] if hand=='left' else settings['reverse'][1]
usable=settings['usable'][0] if hand=='left' else settings['usable'][1]
if return_list:
return settings['person'],dr,rotAngle,reverse,usable
return settings,dr
def subject_settings_v3_4(index,hand,return_list=True):
"""
Subject info for experiment v3.4. Audio no training. Twoples refer to the left and right subject
hands.
2018-02-16
Parameters
----------
index : int
hand : str
Subject hand used.
return_list : bool,True
Returns
-------
settings : dict
dr : str
rotAngle : float
reverse : bool
usable : bool
"""
settings = [{'person':'Subject01_3_4',
'trials':['avatar'],
'reverse':[False,False],
'usable':[False,True]},
{'person':'Subject02_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject03_3_4',
'trials':['avatar'],
'reverse':[True,False],
'usable':[False,True]},
{'person':'Subject04_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject05_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject06_3_4',
'trials':['avatar'],
'reverse':[True,False],
'usable':[False,True]},
{'person':'Subject07_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject08_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject09_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject10_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject11_3_4',
'trials':['avatar'],
'reverse':[True,False],
'usable':[False,True]},
{'person':'Subject12_3_4',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]}
][index]
dr = '../data/UE4_Experiments/%s/%s'%(settings['person'],hand)
try:
rotAngle = pickle.load(open('%s/%s'%(dr,'gpr.p'),'rb'),encoding='latin1')['rotAngle']
except IOError:
rotAngle=np.nan
reverse=settings['reverse'][0] if hand=='left' else settings['reverse'][1]
usable=settings['usable'][0] if hand=='left' else settings['usable'][1]
if return_list:
return settings['person'],dr,rotAngle,reverse,usable
return settings,dr
def subject_settings_v3_5(index,hand,return_list=True):
"""
Subject info for experiment v3.5. Twoples refer to the left and right subject hands.
2018-04-15
Parameters
----------
index : int
hand : str
Subject hand used.
return_list : bool,True
Returns
-------
settings : dict
dr : str
"""
settings = [{'person':'Subject01_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject02_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject03_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject04_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[False,True]},
{'person':'Subject05_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject06_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject07_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject08_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject09_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[False,True]},
{'person':'Subject10_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject11_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject12_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject13_3_5',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject14_3_5',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]}
][index]
dr = '../data/UE4_Experiments/%s/%s'%(settings['person'],hand)
try:
rotAngle = pickle.load(open('%s/%s'%(dr,'gpr.p'),'rb'),encoding='latin1')['rotAngle']
# In the case where the final save in HandSyncExperiment.run_vr did not complete, the
# rotAngle will be a list.
if type(rotAngle) is list:
rotAngle=rotAngle[0] if hand=='left' else rotAngle[1]
except KeyError:
from .experiment import HandSyncExperiment
# Need to reload data from file.
f=[f for f in os.listdir('%s'%dr) if 'an_port_cal' in f]
f.sort()
rotAngle=HandSyncExperiment.read_cal('%s/%s'%(dr,f[-1]),.3)
rotAngle=rotAngle[0] if hand=='left' else rotAngle[1]
except IOError:
rotAngle=np.nan
reverse=settings['reverse'][0] if hand=='left' else settings['reverse'][1]
usable=settings['usable'][0] if hand=='left' else settings['usable'][1]
if return_list:
return settings['person'],dr,rotAngle,reverse,usable
return settings,dr
def subject_settings_v3_6(index,hand,return_list=True):
"""
Subject info for experiment v3.3. Twoples refer to the left and right subject hands.
2018-04-22
Parameters
----------
index : int
hand : str
Subject hand used.
return_list : bool,True
Returns
-------
settings : dict
dr : str
"""
assert hand in ['left','right']
settings = [{'person':'Subject01_3_6',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject02_3_6',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject03_3_6',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject04_3_6',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject05_3_6',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,False]},
{'person':'Subject06_3_6',
'trials':['avatar'],
'reverse':[False,True],
'usable':[True,True]},
{'person':'Subject07_3_6',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]},
{'person':'Subject08_3_6',
'trials':['avatar'],
'reverse':[True,False],
'usable':[False,False]},
{'person':'Subject09_3_6',
'trials':['avatar'],
'reverse':[True,False],
'usable':[True,True]}
][index]
dr = '../data/UE4_Experiments/%s/%s'%(settings['person'],hand)
try:
rotAngle = pickle.load(open('%s/%s'%(dr,'gpr.p'),'rb'),encoding='latin1')['rotAngle']
except IOError:
rotAngle=np.nan
reverse=settings['reverse'][0] if hand=='left' else settings['reverse'][1]
usable=settings['usable'][0] if hand=='left' else settings['usable'][1]
if return_list:
return settings['person'],dr,rotAngle,reverse,usable
return settings,dr
# ------------------ #
# Class definitions. #
# ------------------ #
class VRTrial3_1(object):
def __init__(self,person,modelhandedness,rotation,dr,
fname='trial_dictionaries.p',
reverse=False,
retrain=True):
"""
Parameters
----------
person : str
modelhandedness : list of str
rotation : list of float
dr : str
Members
-------
person : str
modelhandedness : list of str
rotation : float
dr : str
subjectTrial : dict
Full Axis Neuron trial data labeled by part+'T' part+'V'.
templateTrial : dict
Full MotionBuilder trial data labeled by part+'T' part+'V'.
timeSplitTrials
subjectSplitTrials
templateSplitTrials
Methods
-------
info
subject_by_window_dur
subject_by_window_spec
pickle_trial_dicts
pickle_phase
_fetch_windowspec_indices
"""
self.person = person
self.modelhandedness = modelhandedness
self.rotation = rotation
self.dr = dr
self.reverse = reverse
# Load gpr data points.
savedData = pickle.load(open('%s/%s'%(self.dr,'gpr.p'),'rb'),encoding='latin1')
self.gprmodel = savedData['gprmodel']
self.pause=savedData['pause'],savedData['unpause']
self.trialTypes = ['avatar']
try:
data = pickle.load(open('%s/%s'%(self.dr,fname),'rb'))
except Exception:
self.pickle_trial_dicts(1)
data = pickle.load(open('%s/%s'%(self.dr,fname),'rb'))
self.templateTrial = data['templateTrial']
self.subjectTrial = data['subjectTrial']
self.timeSplitTrials = data['timeSplitTrials']
self.templateSplitTrials = data['templateSplitTrials']
self.subjectSplitTrials = data['subjectSplitTrials']
self.windowsByPart = data['windowsByPart']
if retrain:
self.retrain_gprmodel()
def info(self):
print("Person %s"%self.person)
print("Trials available:")
for part in self.trialTypes:
print("%s\tInvisible\tTotal"%part)
for spec,_ in self.windowsByPart[part]:
print("\t%1.2f\t\t%1.2f"%(spec[0],spec[1]))
def subject_by_window_dur(self,windowDur,part):
"""
Params:
-------
windowDur (list)
Duration of visible/invisible cycle.
part (str)
Body part to return.
Returns:
--------
selection (list)
List of trials that have given window duration. Each tuple in list is a tuple of the
( (invisible,total window), time, extracted velocity data ).
"""
raise NotImplementedError("Needs to be fixed.")
ix = []
i=0
for spec,_ in self.windowsByPart[part]:
if np.isclose(windowDur,spec[1]):
ix.append(i)
i += 1
selection = []
for i in ix:
selection.append(( self.windowsByPart[part][i][0],
self.timeSplitTrials[part][i],
self.subjectSplitTrials[part][i] ))
return selection
def template_by_window_dur(self,windowDur,part):
raise NotImplementedError("Needs to be fixed.")
ix = []
i=0
for spec,_ in self.windowsByPart[part]:
if np.isclose(windowDur,spec[1]):
ix.append(i)
i += 1
selection = []
for i in ix:
selection.append(( self.windowsByPart[part][i][0],
self.timeSplitTrials[part][i],
self.templateSplitTrials[part][i] ))
return selection
def subject_by_invisible_dur(self,windowDur,part):
raise NotImplementedError("Needs to be fixed.")
ix = []
i=0
for spec,_ in self.windowsByPart[part]:
if np.isclose(windowDur,spec[0]):
ix.append(i)
i += 1
selection = []
for i in ix:
selection.append(( self.windowsByPart[part][i][0],
self.timeSplitTrials[part][i],
self.subjectSplitTrials[part][i] ))
return selection
def subject_by_window_spec(self,windowSpec,trial_type,precision=None):
"""Automatically search through left and right hand trials."""
selection = []
for specix,spec in enumerate(windowSpec):
ix = self._fetch_windowspec_indices([spec],trial_type,precision=precision)
for ix_ in ix:
selection.append(( self.windowsByPart[trial_type][ix_][0],
self.timeSplitTrials[trial_type][ix_],
self.subjectSplitTrials[trial_type][ix_].copy() ))
# Iterate also through hand0 or avatar0, which contains the other hand..
#if trial_type.isalpha():
# selection += self.subject_by_window_spec([windowSpec[specix]],
# trial_type+'0',
# precision=precision)
return selection
def template_by_window_spec(self,windowSpec,trial_type,precision=None):
"""Automatically search through left and right hand trials."""
selection = []
for specix,spec in enumerate(windowSpec):
ix = self._fetch_windowspec_indices([spec],trial_type,precision=precision)
for ix_ in ix:
selection.append(( self.windowsByPart[trial_type][ix_][0],
self.timeSplitTrials[trial_type][ix_],
self.templateSplitTrials[trial_type][ix_].copy() ))
# Iterate also through hand0 or avatar0, which contains the other hand..
#if trial_type.isalpha():
# selection += self.template_by_window_spec([windowSpec[specix]],
# trial_type+'0',
# precision=precision)
return selection
def template_by_invisible_dur(self,windowDur,part):
raise NotImplementedError("Needs to be fixed.")
ix = []
i=0
for spec,_ in self.windowsByPart[part]:
if np.isclose(windowDur,spec[0]):
ix.append(i)
i += 1
selection = []
for i in ix:
selection.append(( self.windowsByPart[part][i][0],
self.timeSplitTrials[part][i],
self.templateSplitTrials[part][i] ))
if trialType.isalpha():
return selection + self.template_by_invisible_dur(windowSpec,trialType+'0')
return selection
def visibility_by_window_spec(self,windowSpec,trial_type,precision=None):
selection = []
for specix,spec in enumerate(windowSpec):
ix = self._fetch_windowspec_indices([spec],trial_type,precision=precision)
if len(ix)>0:
selection.append(( self.windowsByPart[trial_type][ix[0]][0],
self.timeSplitTrials[trial_type][ix[0]],
self.templateSplitTrials[trial_type+'visibility'][ix[0]] ))
# Iterate also through hand0 or avatar0, which contains the other hand..
#if trial_type.isalpha():
# selection += self.visibility_by_window_spec([windowSpec[specix]],
# trial_type+'0',
# precision=precision)
return selection
def phase_by_window_dur(self,source,windowDur,trialType):
"""
Return instantaneous phase from bandpass filtered velocities on trial specificied by window
duration.
Params:
-------
source (str)
windowDur (list of floats)
trialType (str)
'avatar', 'avatar0', 'hand', 'hand0'
"""
raise NotImplementedError("Needs to be fixed.")
ix = []
i = 0
for spec,_ in self.windowsByPart[trialType]:
if np.isclose(windowDur,spec[1]):
ix.append(i)
i += 1
selection = []
for i in ix:
try:
if source=='subject' or source=='s':
phases = pickle.load(open('%s/subject_phase_%s_%d.p'%(self.dr,trialType,i),
'rb'))['phases']
elif source=='template' or source=='t':
phases = pickle.load(open('%s/template_phase_%s_%d.p'%(self.dr,trialType,i),
'rb'))['phases']
phases = [np.vstack(p) for p in phases]
selection.append(( self.windowsByPart[trialType][i][0],phases ))
except IOError:
print("Trial %d in trial type %s not found."%(i,trialType))
return selection
def phase_by_window_spec(self,source,windowSpec,trial_type):
"""
Parameters
----------
source : str
windowSpec : list
trial_type : str
"""
selection = []
for specix,spec in enumerate(windowSpec):
ix = self._fetch_windowspec_indices([spec],trial_type,precision=precision)
if len(ix)>0:
selection.append(( self.windowsByPart[trial_type][ix[0]][0],
self.timeSplitTrials[trial_type][ix[0]],
self.templateSplitTrials[trial_type][ix[0]] ))
try:
if source=='subject' or source=='s':
data = pickle.load(open('%s/subject_phase_%s_%d.p'%(self.dr,trial_type,ix[0]),'rb'))
phases,vs = data['phases'],data['vs']
elif source=='template' or source=='t':
data = pickle.load(open('%s/template_phase_%s_%d.p'%(self.dr,trial_type,ix[0]),'rb'))
phases,vs = data['phases'],data['vs']
phases = [np.vstack(p) for p in phases]
selection.append(( self.windowsByPart[trial_type][ix[0]][0],phases ))
except IOError:
print("Trial %d in trial type %s not found."%(ix[0],trial_type))
# Iterate also through hand0 or avatar0, which contains the other hand.
if trial_type.isalpha():
selection += self.phase_by_window_spec(source,
[windowSpec[specix]],
trial_type+'0',
precision=precision)
return selection
def filtv_by_window_spec(self,source,windowSpec,trialType,search_all=True):
"""
Returns:
--------
list of twoples (windowSpec, filtv) where filtv is a list of 3 arrays corresponding to each dimension
"""
raise NotImplementedError()
ix = self._fetch_windowspec_indices(windowSpec,trialType,precision=precision)
selection = []
for i in ix:
if source=='subject' or source=='s':
data = pickle.load(open('%s/subject_phase_%s_%d.p'%(self.dr,trialType,i),'rb'))
phases,vs = data['phases'],data['vs']
elif source=='template' or source=='t':
data = pickle.load(open('%s/template_phase_%s_%d.p'%(self.dr,trialType,i),'rb'))
phases,vs = data['phases'],data['vs']
else:
raise Exception
vs = [np.vstack(p) for p in vs]
selection.append(( self.windowsByPart[trialType][i][0],vs ))
if trialType.isalpha() and search_all:
return selection + self.filtv_by_window_spec(source,windowSpec,trialType+'0',False)
return selection
def dphase_by_window_dur(self,windowDur,trialType):
"""
Difference in phase between subject and template motion.
"""
raise NotImplementedError
from misc.angle import mod_angle
subjectPhase = self.phase_by_window_dur('s',windowDur,trialType)
templatePhase = self.phase_by_window_dur('t',windowDur,trialType)
dphase = []
for i in range(len(subjectPhase)):
dphase.append(( subjectPhase[i][0],
[mod_angle( s-t ) for s,t in zip(subjectPhase[i][1],templatePhase[i][1])] ))
if trialType.isalpha():
return dphase + self.dphase_by_window_dur(windowDur,trialType+'0')
return dphase
def dphase_by_window_spec(self,windowSpec,trialType):
"""
Difference in phase between subject and template motion.
"""
from misc.angle import mod_angle
subjectPhase = self.phase_by_window_spec('s',windowSpec,trialType)
templatePhase = self.phase_by_window_spec('t',windowSpec,trialType)
dphase = []
for i in range(len(subjectPhase)):
dphase.append(( subjectPhase[i][0],
[mod_angle( s-t ) for s,t in zip(subjectPhase[i][1],templatePhase[i][1])] ))
return dphase
def retrain_gprmodel(self,**gpr_kwargs):
"""Train gprmodel again. This is usually necessary when the GPR class is modified and the performance
values need to be calculated again.
If available, use precoputed DTW with cost function to keep subject trajectory within
bounds.
Parameters
----------
**gpr_kwargs
"""
print("Retraining model...")
from .coherence import DTWPerformance,GPREllipsoid
perfEval=DTWPerformance(dt_threshold=.68)
gprmodel=GPREllipsoid(tmin=self.gprmodel.tmin,tmax=self.gprmodel.tmax,
fmin=self.gprmodel.fmin,fmax=self.gprmodel.fmax,
mean_performance=self.gprmodel.performanceData.mean(),
**gpr_kwargs)
p=np.zeros(len(self.timeSplitTrials['avatar']))
# Try to load DTW alignment path that would have been calculated with regularization.
version=self.person[-3:]
homedr=os.path.expanduser('~')
f=homedr+'/Dropbox/Research/tango/py/cache/dtw_v%s.p'%version
if os.path.isfile(f):
print("Using cached DTW path file.")
pathList=pickle.load(open(f,'rb'))['pathList'][self._find_subject_settings_index()]
assert len(self.templateSplitTrials['avatar'])==len(pathList)
for i,(t,sv,avv,path) in enumerate(zip(self.timeSplitTrials['avatar'],
self.subjectSplitTrials['avatar'],
self.templateSplitTrials['avatar'],
pathList)):
p[i]=perfEval.time_average_binary(avv[:,1:],sv[:,1:],
dt=1/30,#t[1]-t[0],
path=path)
else:
for i,(t,sv,avv) in enumerate(zip(self.timeSplitTrials['avatar'],
self.subjectSplitTrials['avatar'],
self.templateSplitTrials['avatar'])):
p[i]=perfEval.time_average_binary(avv[:,1:],sv[:,1:],dt=t[1]-t[0],bds=[1,t.max()-1])
assert ((1>p)&(p>0)).all()
gprmodel.update(self.gprmodel.ilogistic(p),self.gprmodel.durations,self.gprmodel.fractions)
self.gprmodel=gprmodel
def _find_subject_settings_index(self):
"""Find where this subject and hand would be located in the flat list of all subject
settings. This is bit hacked.
Returns
-------
index : int
"""
done=False
if self.person[-3:]=='3_3':
subject_settings=subject_settings_v3_3
elif self.person[-3:]=='3_4':
subject_settings=subject_settings_v3_4
elif self.person[-3:]=='3_5':
subject_settings=subject_settings_v3_5
elif self.person[-3:]=='3_6':
subject_settings=subject_settings_v3_6
else:
raise Exception("Unrecognized experiment version.")
totalCounter=0
subjectCounter=0
while not done:
try:
person,dr,rotAngle,reverse,usable=subject_settings(subjectCounter,'left')
if self.person==person and self.modelhandedness[0].lower()=='right':
return totalCounter
elif usable:
totalCounter+=1
person,dr,rotAngle,reverse,usable=subject_settings(subjectCounter,'right')
if self.person==person and self.modelhandedness[0].lower()=='left':
return totalCounter
elif usable:
totalCounter+=1
subjectCounter+=1
except IndexError:
done=True
raise Exception("Subject not found in list: %s"%self.person)
def pickle_trial_dicts(self,disp=False):
"""
Put data for analysis into easily accessible pickles. Right now, I extract only visibility and hand
velocities for AN port data and avatar's motionbuilder files.
Parameters
----------
disp : bool,False
"""
from .axis_neuron import extract_AN_port
from .pipeline import extract_motionbuilder_model3_3
from .utils import match_time
from .ue4 import load_visibility
import dill as pickle
from .experiment import remove_pause_intervals
# Load AN data.
df = pickle.load(open('%s/%s'%(self.dr,'quickload_an_port_vr.p'),'rb'))['df']
windowsByPart,_,_ = self.window_specs(self.person,self.dr)
# Sort trials into the hand, arm, and avatar trial dictionaries: subjectTrial, templateTrial,
# hmdTrials. These contain arrays for time that were interpolated in for regular sampling and
# functions for velocities.
subjectTrial,templateTrial,hmdTrials = {},{},{}
timeSplitTrials,subjectSplitTrials,templateSplitTrials = {},{},{}
for trialno,part in enumerate(self.trialTypes):
if disp:
print("Processing %s..."%part)
# Load visibility time points saved by UE4 and remove pause intervals.
if part.isalpha():
visible,invisible = load_visibility(part+'_visibility',self.dr)
else:
visible,invisible = load_visibility(part[:-1]+'_visibility_0',self.dr)
visible,_=remove_pause_intervals(visible.tolist(),list(zip(*self.pause)))
invisible,_=remove_pause_intervals(invisible.tolist(),list(zip(*self.pause)))
visible,invisible=np.array(visible),np.array(invisible)
# Start and end times counting only the time the simulation is running (and not paused).
exptStartEnd = [visible[0],invisible[-1]]
# Extract template.
mbV,mbT = extract_motionbuilder_model3_3( self.modelhandedness[trialno],
reverse_time=self.reverse )
showIx = mbT < (exptStartEnd[1]-exptStartEnd[0]).total_seconds()
templateTrial[part+'T'] = mbT[showIx][::2]
templateTrial[part+'V'] = mbV
# Extract subject from port file.
anT,anX,anV,anA = extract_AN_port( df,self.modelhandedness[trialno],
rotation_angle=self.rotation )
# Remove pauses.
anT,_,removeIx=remove_pause_intervals(anT.tolist(),list(zip(*self.pause)),True)
anT=np.array(anT)
anV=np.delete(anV[0],removeIx,axis=0)
# Remove parts that extend beyond trial.
showIx = (anT>=exptStartEnd[0]) & (anT<=exptStartEnd[1])
anT,anV = anT[showIx],anV[showIx]
# Save into variables used here.
subjectTrial[part+'T'],subjectTrial[part+'V'] = anT,anV
# Put trajectories on the same time samples so we can pipeline our regular computation.
offset = (subjectTrial[part+'T'][0]-exptStartEnd[0]).total_seconds()
subjectTrial[part+'V'],subjectTrial[part+'T'] = match_time(subjectTrial[part+'V'],
subjectTrial[part+'T'],
1/30,
offset=offset,
use_univariate=True)
# Separate the different visible trials into separate arrays.
# Times for when visible/invisible windows start.
start = np.zeros((len(visible)+len(invisible)),dtype=object)
start[::2] = visible
start[1::2] = invisible
# Units of seconds.
start = np.array([t.total_seconds() for t in np.diff(start)])
start = np.cumsum(start)
invisibleStart = start[::2] # as seconds
visibleStart = start[1::2] # as seconds
# When target is invisible, set visibility to 0.
visibility = np.ones_like(templateTrial[part+'T'])
for i,j in zip(invisibleStart,visibleStart):
if i<j:
visibility[(templateTrial[part+'T']>=i) & (templateTrial[part+'T']<j)] = 0
if len(visible)<len(invisible):
visibility[(templateTrial[part+'T']>=invisible[-1])] = 0
templateTrial[part+'visibility'] = visibility
# Separate single data take into separate trials.
timeSplitTrials[part],subjectSplitTrials[part],templateSplitTrials[part] = [],[],[]
templateSplitTrials[part+'visibility'] = []
for spec,startendt in windowsByPart[part]:
startendt = ((startendt[0]-exptStartEnd[0]).total_seconds(),
(startendt[1]-exptStartEnd[0]).total_seconds())
# Save time.
timeix = (templateTrial[part+'T']<=startendt[1])&(templateTrial[part+'T']>=startendt[0])
t = templateTrial[part+'T'][timeix]
timeSplitTrials[part].append(t)
# Save visibility window.
templateSplitTrials[part+'visibility'].append( visibility[timeix] )
# Save velocities.
templateSplitTrials[part].append( templateTrial[part+'V'](t) )
subjectSplitTrials[part].append( subjectTrial[part+'V'](t) )
pickle.dump({'templateTrial':templateTrial,
'subjectTrial':subjectTrial,