-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathrun_conversion.py
1010 lines (832 loc) · 43 KB
/
run_conversion.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
# -*- coding: utf-8 -*-
__authors__ = 'Sol', 'r-zemblys'
### CONVERSION SCRIPT SETTINGS ###
INCLUDE_TRACKERS = (
'dpi',
'eyefollower',
'eyelink',
'eyetribe',
'hispeed1250',
'hispeed240',
'red250',
'red500',
'redm',
't60xl',
'tx300',
'x2'
)
MONOCULAR_TRACKERS = (
'dpi',
# 'eyefollower',
'eyelink',
# 'eyetribe',
'hispeed1250',
'hispeed240',
# 'red250',
# 'red500',
'redm',
# 't60xl',
# 'tx300',
# 'x2'
)
INCLUDE_SUB = 'ALL'
INPUT_FILE_ROOT = r"/media/Data/EDQ/data_hdf5"
OUTPUT_FOLDER = r'/media/Data/EDQ/data_npy'
SAVE_NPY = True
SAVE_TXT = False
#DPICAL is required to calibrate DPI
BLOCKS_TO_EXPORT = ['DPICAL', 'FS']
CALIBRATE_DPI = True
#Check for binocular data averaging
CHECK_BDA = True
DATASET = 'EDQ_LUND'
PLOT_TRACKERS = (
'dpi',
'eyefollower',
'eyelink',
'eyetribe',
'hispeed1250',
'hispeed240',
'red250',
'red500',
'redm',
't60xl',
'tx300',
'x2'
)
plot_multisession = True
GLOB_PATH_PATTERN = INPUT_FILE_ROOT + r"/*/*/*.hdf5"
##################################
import os, sys
import glob
import re
from timeit import default_timer as getTime
import numpy as np
import matplotlib.pylab as plt
plt.ion()
import tables
from collections import OrderedDict
from constants import (MONOCULAR_EYE_SAMPLE, BINOCULAR_EYE_SAMPLE, MESSAGE,
MULTI_CHANNEL_ANALOG_INPUT,
wide_row_dtype, msg_txt_mappings, et_mappings,
dpi_cal_fix, stim_pos_mappings,
smc_dtype
)
from edq_shared import (getFullOutputFolderPath, nabs,
save_as_txt, parseTrackerMode, VisualAngleCalc,
filter_trackloss,
plot_data,
)
if 'dpi' in INCLUDE_TRACKERS:
import cv2 #Bilateral filter from OpenCV library is used to filter DPI data
from edq_shared import detect_rollingWin, filter_bilateral
try:
from yaml import load, dump
from yaml import CLoader as Loader, CDumper as Dumper
except ImportError:
from yaml import Loader, Dumper
if sys.version_info[0] != 2 or sys.version_info[1] >= 7:
def construct_yaml_unistr(self, node):
return self.construct_scalar(node)
Loader.add_constructor(u'tag:yaml.org,2002:str', construct_yaml_unistr)
####
### Conversion variables
binoc_sample_fields = ['session_id', 'device_time', 'time',
'left_gaze_x', 'left_gaze_y', 'left_pupil_measure1',
'right_gaze_x', 'right_gaze_y', 'right_pupil_measure1',
'status']
LEFT_EYE_POS_X_IX = binoc_sample_fields.index('left_gaze_x')
LEFT_EYE_POS_Y_IX = binoc_sample_fields.index('left_gaze_y')
RIGHT_EYE_POS_X_IX = binoc_sample_fields.index('right_gaze_x')
RIGHT_EYE_POS_Y_IX = binoc_sample_fields.index('right_gaze_y')
mono_sample_fields = ['session_id', 'device_time', 'time',
'gaze_x', 'gaze_y', 'pupil_measure1',
'gaze_x', 'gaze_y', 'pupil_measure1',
'status']
dpi_sample_fields = ['session_id', 'device_time', 'time',
'AI_4', 'AI_5', 'device_id',
'AI_0', 'AI_1', 'device_id',
'AI_2']
screen_measure_fields = ('screen_width', 'screen_height', 'eye_distance')
cv_fields = ['SESSION_ID', 'trial_id', 'TRIAL_START', 'TRIAL_END', 'posx',
'posy', 'dt', 'ROW_INDEX', 'BLOCK']
TARGET_POS_X_IX = cv_fields.index('posx')
TARGET_POS_Y_IX = cv_fields.index('posy')
FIX_DPI_CAL = False
special_multisession_cases = []
###
def getInfoFromPath(fpath):
"""
:param fpath:
:return:
"""
if fpath.lower().endswith(".hdf5"):
fpath, fname = os.path.split(fpath)
return fpath.rsplit(os.path.sep, 3)[-2], np.uint(
re.split('_|.hdf5', fname)[-2])
def analyseit(fpath):
"""
:param fpath:
:return:
"""
tracker_type, sub = getInfoFromPath(fpath)
if INCLUDE_SUB == 'ALL':
return (tracker_type in INCLUDE_TRACKERS)
else:
return (tracker_type in INCLUDE_TRACKERS) & (sub in INCLUDE_SUB)
def openHubFile(filepath, filename, mode):
"""
Open an HDF5 DataStore file.
"""
hubFile = tables.openFile(os.path.join(filepath, filename), mode)
return hubFile
def getEventTableForID(hub_file, event_type):
"""
Return the pytables event table for the given EventConstant event type
:param hub_file: pytables hdf5 file
:param event_type: int
:return:
"""
evt_table_mapping = hub_file.root.class_table_mapping.read_where(
'class_id == %d' % (event_type))
return hub_file.getNode(evt_table_mapping[0]['table_path'])
def num(s):
"""
:param s:
:return:
"""
try:
return int(s)
except ValueError:
try:
return float(s)
except ValueError:
return s
def getSessionDataFromMsgEvents(hub_file):
"""
Get all the session data that was saved as messages and not in the
session meta data table
:param hub_file:
:return: dict
"""
# and not in the session meta data table
msg_table = getEventTableForID(hub_file, MESSAGE)
# < 20 msg's written in this exp, so just read them all
session_ids = np.unique(msg_table.read()['session_id'])
session_infos = OrderedDict()
for sid in session_ids:
session_info = OrderedDict()
session_infos[sid] = session_info
msg_event_text = msg_table.read_where("session_id == %d" % (sid))[
'text']
_msg_event_dict = dict()
for msg in msg_event_text:
msplit = msg.split(':')
_msg_event_dict[msplit[0].strip()] = [t.strip() for t in msplit[1:]]
if msg.find('ioHub Experiment started') > -1:
_msg_event_dict['exp_date'] = [msg.strip('ioHub Experiment started')]
# Parse out (painfully) the data of interest
for org_title, txt_title in msg_txt_mappings.iteritems():
msg_data = _msg_event_dict.get(org_title)
if msg_data:
if len(msg_data) == 1:
msg_data = msg_data[0]
session_info[txt_title] = num(msg_data)
elif org_title == 'Stimulus Screen ID':
pstr = msg_data[1].split(',')[0:2]
session_info[txt_title] = num(pstr[0][1:])
session_info[msg_txt_mappings['Stimulus Screen ID2']] = num(
pstr[1][:-1])
return session_infos
def convertEDQ(hub_file, screen_measures, et_model):
"""
:param hub_file:
:param screen_measures:
:param et_model:
:return:
"""
display_size_mm = screen_measures['screen_width'], screen_measures[
'screen_height']
sample_data_by_session = []
session_info_dict = getSessionDataFromMsgEvents(hub_file)
if CALIBRATE_DPI & FIX_DPI_CAL & (et_model == 'dpi'):
print 'Warning! ROW_INDEX fix for DPI calibration is ENABLED'
for session_id, session_info in session_info_dict.items():
# Get the condition variable set rows for the 'FS' and/or 'DPICAL' trial type
for block in BLOCKS_TO_EXPORT:
ecvTable = hub_file.root.data_collection.condition_variables\
.EXP_CV_1
cv_rows = ecvTable.read_where(
'(BLOCK == "%s") & (SESSION_ID == %d)' % (block, session_id))
cv_row_count = len(cv_rows)
if cv_row_count == 0:
# print "Skipping Session %d, not FS blocks" % (session_id)
continue
display_size_pix = session_info['display_width_pix'], session_info[
'display_height_pix']
pix2deg = VisualAngleCalc(display_size_mm, display_size_pix,
screen_measures['eye_distance']).pix2deg
session_info.update(screen_measures)
session_info_vals = session_info.values()
tracking_eye = session_info['eyetracker_mode']
# Get the eye sample table
if tracking_eye == 'Binocular':
sample_table = getEventTableForID(hub_file,
BINOCULAR_EYE_SAMPLE)
if sample_table.nrows == 0:
sample_table = getEventTableForID(hub_file,
MONOCULAR_EYE_SAMPLE)
sample_fields = mono_sample_fields
else:
sample_fields = binoc_sample_fields
else:
if et_model == 'dpi':
sample_table = getEventTableForID(hub_file, MULTI_CHANNEL_ANALOG_INPUT)
sample_fields = dpi_sample_fields
else:
sample_table = getEventTableForID(hub_file, MONOCULAR_EYE_SAMPLE)
if sample_table.nrows == 0:
sample_table = getEventTableForID(hub_file, BINOCULAR_EYE_SAMPLE)
sample_fields = binoc_sample_fields
else:
sample_fields = mono_sample_fields
if et_model == 'eyetribe':
# Use raw_x, raw_y instead of gaze
sample_fields = [s.replace('gaze', 'raw') for s in
sample_fields]
# Data collected for eyetribe seems to have been using a
# version of
# script
# that calculated the time incorrectly; so here we fix it.
delay_col = sample_table.col('delay')[0]
if delay_col != 0.0:
# fix the time and delay fields of eye tribe files;
# changes are
# saved back t hdf5
time_mod_count = sample_table.modify_column(0,
sample_table.nrows,
column=sample_table.col(
'logged_time'),
colname='time')
delay_nod_count = sample_table.modify_column(0,
sample_table.nrows,
column=sample_table.col(
'left_gaze_z'),
colname='delay')
# create wide format txt output
trial_end_col_index = cv_fields.index('TRIAL_END')
sample_array_list = []
for row_index, cv_set in enumerate(cv_rows[:-1]):
assert session_id == cv_set['SESSION_ID']
next_cvs = cv_rows[row_index + 1]
# Get current condition var value str. Since sample time period
# selection is between cv_set['TRIAL_START'], next_cvs[
# 'TRIAL_START']
# set the TRIAL_END var for current row to == next_cvs[
# 'TRIAL_START']
# for targets 0 -(n-1)
cv_vals = [cv_set[cvf] for cvf in cv_fields]
####Fixes ROW_INDEX in DPI calibration routine for 2014 Apr-May EDQ recordings
if FIX_DPI_CAL & (cv_vals[-1] == 'DPICAL'):
cv_vals[-2]=dpi_cal_fix[cv_vals[-2]]
###
tpdegxy = pix2deg(cv_vals[TARGET_POS_X_IX],
cv_vals[TARGET_POS_Y_IX])
cv_vals[trial_end_col_index] = next_cvs['TRIAL_START']
targ_pos_samples = sample_table.where(
"(session_id == %d) & (time >= %.6f) & (time <= %.6f)" % (
cv_set['SESSION_ID'], cv_set['TRIAL_START'],
next_cvs['TRIAL_START']))
for sample in targ_pos_samples:
sample_vals = [sample[svn] for svn in sample_fields]
if et_model == 'dpi':
rdegxy = (sample_vals[RIGHT_EYE_POS_X_IX],
sample_vals[RIGHT_EYE_POS_Y_IX])
ldegxy = (sample_vals[LEFT_EYE_POS_X_IX],
sample_vals[LEFT_EYE_POS_Y_IX])
else:
rdegxy = pix2deg(sample_vals[RIGHT_EYE_POS_X_IX],
sample_vals[RIGHT_EYE_POS_Y_IX])
ldegxy = pix2deg(sample_vals[LEFT_EYE_POS_X_IX],
sample_vals[LEFT_EYE_POS_Y_IX])
try:
sample_array_list.append(tuple(
session_info_vals + cv_vals + sample_vals + list(
tpdegxy) + list(
ldegxy) + list(rdegxy)))
except:
import traceback
traceback.print_exc()
# process last target pos.
cv_set = cv_rows[-1]
cv_vals = [cv_set[cvf] for cvf in cv_fields]
tpdegxy = pix2deg(cv_vals[TARGET_POS_X_IX],
cv_vals[TARGET_POS_Y_IX])
targ_pos_samples = sample_table.where(
"(session_id == %d) & (time >= %.6f) & (time <= %.6f)" % (
cv_set['SESSION_ID'], cv_set['TRIAL_START'],
cv_set['TRIAL_END']))
for sample in targ_pos_samples:
sample_vals = [sample[svn] for svn in sample_fields]
if et_model == 'dpi':
rdegxy = (sample_vals[RIGHT_EYE_POS_X_IX],
sample_vals[RIGHT_EYE_POS_Y_IX])
ldegxy = (sample_vals[LEFT_EYE_POS_X_IX],
sample_vals[LEFT_EYE_POS_Y_IX])
else:
rdegxy = pix2deg(sample_vals[RIGHT_EYE_POS_X_IX],
sample_vals[RIGHT_EYE_POS_Y_IX])
ldegxy = pix2deg(sample_vals[LEFT_EYE_POS_X_IX],
sample_vals[LEFT_EYE_POS_Y_IX])
try:
sample_array_list.append(tuple(
session_info_vals + cv_vals + sample_vals + list(
tpdegxy) + list(ldegxy) + list(
rdegxy)))
except:
import traceback
traceback.print_exc()
sample_data_by_session.append(sample_array_list)
return sample_data_by_session
def getScreenMeasurements(dpath, et_model, et_model_display_configs):
"""
:param dpath:
:param et_model:
:param et_model_display_configs:
:return:
"""
# et_model, _ = getInfoFromPath(dpath)
display_param = et_model_display_configs.get(et_model)
if display_param is None:
et_config_path = glob.glob('./configs/*%s.yaml' % (et_model))
if et_config_path:
et_config_path = nabs(et_config_path[0])
display_config = load(file(et_config_path, 'r'), Loader=Loader)
dev_list = display_config.get('monitor_devices')
for d in dev_list:
if d.keys()[0] == 'Display':
d = d['Display']
width = d.get('physical_dimensions', {}).get('width')
height = d.get('physical_dimensions', {}).get('height')
eye_dist = d.get('default_eye_distance', {}).get(
'surface_center')
et_model_display_configs[et_model] = OrderedDict()
et_model_display_configs[et_model][
screen_measure_fields[0]] = width
et_model_display_configs[et_model][
screen_measure_fields[1]] = height
et_model_display_configs[et_model][
screen_measure_fields[2]] = eye_dist
return et_model_display_configs[et_model]
return display_param
def checkFileIntegrity(hub_file):
"""
:param hub_file:
:return:
"""
try:
tm = hub_file.root.class_table_mapping
except:
print "\n>>>>>>\nERROR processing Hdf5 file: %s\n\tFile does not have " \
"" \
"a root.class_table_mapping table.\n\tSKIPPING FILE.\n<<<<<<\n" \
% (
file_path)
if hub_file:
hub_file.close()
hub_file = None
return False
try:
tm = hub_file.root.data_collection.condition_variables.EXP_CV_1
except:
print "\n>>>>>>\nERROR processing Hdf5 file: %s\n\tFile does not have " \
"" \
"a root.data_collection.condition_variables.EXP_CV_1 " \
"table.\n\tSKIPPING FILE.\n<<<<<<\n" % (
file_path)
if hub_file:
hub_file.close()
hub_file = None
return False
return True
def handle_dpi_multisession(data):
#Finds last block of calibration and Fixate-Saccade task
print 'Multisession DPI data found'
sessions = np.unique(data['session_id'])
_fs = []
_cal = []
for session_id in sessions:
session_data = data['session_id'] == session_id
DATA_SESSION = data[session_data]
cal_block = DATA_SESSION['BLOCK'] == 'DPICAL'
exp_block = DATA_SESSION['BLOCK'] == 'FS'
_cal.append( len(np.unique(DATA_SESSION[cal_block]['ROW_INDEX'])))
_fs.append( len(np.unique(DATA_SESSION[exp_block]['ROW_INDEX'])))
_fs_LastOcc = len(_fs) - 1 - _fs[::-1].index(49)
_cal_=_cal[:_fs_LastOcc+1]
_cal_LastOcc = len(_cal_) - 1 - _cal_[::-1].index(25)
session_data = data['session_id'] == sessions[_cal_LastOcc]
DATA_CAL=data[session_data]
cal_block = DATA_CAL['BLOCK'] == 'DPICAL'
DATA_CAL=DATA_CAL[cal_block]
session_data = data['session_id'] == sessions[_fs_LastOcc]
DATA_EXP=data[session_data]
exp_block = DATA_EXP['BLOCK'] == 'FS'
DATA_EXP=DATA_EXP[exp_block]
return np.hstack((DATA_CAL, DATA_EXP)), sessions[_cal_LastOcc], sessions[_fs_LastOcc]
### Handle multisession END ###
def build_polynomial(X, Y, poly_type):
if poly_type == 'linear':
Px = np.vstack((X, np.ones(len(X)))).T
Py = np.vstack((Y, np.ones(len(Y)))).T
elif poly_type == 'villanueva':
Px = np.vstack((X**2, Y**2, X, Y, X*Y, np.ones(len(X)))).T
Py = np.vstack((X**2, Y**2, X, Y, X*Y, np.ones(len(X)))).T
return Px, Py
#Custom exception
class ConversionError(RuntimeError):
def __init__(self, arg):
self.args = arg
self.message = arg
def nan_equal(a,b):
try:
np.testing.assert_equal(a,b)
except AssertionError:
return False
return True
#Custom conversion settings
if DATASET == 'EDQ_LUND':
'''
In 2014 Apr-May EDQ recordings ROW_INDEX in DPICAL block is not set right.
Set FIX_DPI_CAL to True if dealing with these recordings
'''
FIX_DPI_CAL = True
'''
Manual multisession handler.
Some recordings contain data from several participants
due to mistake when setting subject id.
*special_multisession_cases* array can be used to manualy couple session id
with subject id.
Format:
('subject_id', np.uint8)
('eyetracker_model', str, 32)
('session_id', np.uint8)
'''
special_multisession_cases.append((121, 'eyelink', 1))
special_multisession_cases.append((124, 'eyelink', 2))
special_multisession_cases.append((91, 'red250', 1))
special_multisession_cases.append((95, 'red250', 2))
special_multisession_cases.append((84, 'red500', 1))
special_multisession_cases.append((87, 'red500', 2))
special_multisession_cases=np.array(special_multisession_cases, dtype=smc_dtype)
#DPI calibration config
if 'dpi' in INCLUDE_TRACKERS:
cal_point_sets = dict([
(5, np.array([0,4,12,20,24])),
(9, np.array([0,2,4,10,12,14,20,22,24])),
(14, np.array([1,2,35,7,9,11,12,13,15,17,19,21,22,23])),
(16, np.array([0,6,18,24,4,8,16,20,2,7,17,22,10,11,13,14])),
(17, np.array([0,6,12,18,24,4,8,16,20,2,7,17,22,10,11,13,14])),
(25, np.arange(25))
])
win_select_funcs = dict([
('roll', detect_rollingWin)
])
calibration_settings_set = [{
'poly_type': 'villanueva',
'cal_point_set': 16,
'min_cal_points': 8,
'win_select_func': 'roll',
'win_size': 0.175,
'win_type': 'sample',
'window_skip': 0.2,
'wsa': 'fiona' ,
'units': 'gaze',
}]
if not os.path.exists(OUTPUT_FOLDER):
os.mkdir(OUTPUT_FOLDER)
OUTPUT_FOLDER = getFullOutputFolderPath(OUTPUT_FOLDER)
print 'OUTPUT_FOLDER:', OUTPUT_FOLDER
DATA_FILES = [nabs(fpath) for fpath in glob.glob(GLOB_PATH_PATTERN) if
analyseit(fpath)]
#Check for dublicates
sub_dict = dict()
for et_model in INCLUDE_TRACKERS:
sub_dict[et_model] = []
for file_path in DATA_FILES:
et, sub = getInfoFromPath(file_path)
if et==et_model:
sub_dict[et_model].append(sub)
if len(sub_dict[et_model]) != len(np.unique(sub_dict[et_model])):
unique_ids, unique_counts = np.unique(sub_dict[et_model], return_counts=True)
print 'Dublicate subject ids in %s:'%et_model, \
unique_ids[np.argwhere(unique_counts > 1).flatten()]
sys.exit()
############### MAIN RUNTIME SCRIPT ########################
#
# Below is the actual script that is run when this file is run through
# the python interpreter. The code above defines functions used by the below
# runtime script.
#
if __name__ == '__main__':
et_model_display_configs = dict()
scount = 0
start_time = getTime()
if not os.path.exists(OUTPUT_FOLDER):
os.mkdir(OUTPUT_FOLDER)
col_count = len(wide_row_dtype.names)
format_str = "{}\t" * col_count
format_str = format_str[:-1] + "\n"
row_names = wide_row_dtype.names
header_line = '\t'.join(row_names) + '\n'
file_proc_count = 0
total_file_count = len(DATA_FILES)
hub_file = None
file_log = None
for file_path in DATA_FILES:
file_log = open(OUTPUT_FOLDER + '/conversion.log', 'a')
try:
# if 1:
t0 = getTime()
dpath, dfile = os.path.split(file_path)
print "Processing file %d / %d. \r" % (file_proc_count + 1, total_file_count)
et_model, sub = getInfoFromPath(file_path)
print 'tracker: {et_model}, sub: {sub}'.format(et_model=et_model, sub=sub)
et_dir = nabs(r"%s/%s" % (OUTPUT_FOLDER, et_model))
if not os.path.exists(et_dir):
os.mkdir(et_dir)
if et_model == 'eyetribe':
# open eyetribe files in update mode so time stamp issue can
# be fixed in files.
hub_file = openHubFile(dpath, dfile, 'a')
else:
hub_file = openHubFile(dpath, dfile, 'r')
if not checkFileIntegrity(hub_file):
# file_log.write('[FILE_CORRUPT]\tfile: {file_path}\n'.format(file_path=file_path ))
raise ConversionError(str("FILE_CORRUPT"))
screen_measurments = getScreenMeasurements(file_path, et_model, et_model_display_configs)
wide_format_samples_by_session = convertEDQ(hub_file,
screen_measurments,
et_model
)
if wide_format_samples_by_session == None or len(wide_format_samples_by_session) == 0:
print "\n>>>>>>\nERROR processing Hdf5 file: %s\n\tFile has " \
"no 'FS' BLOCK COND VARS.\n\tSKIPPING FILE.\n<<<<<<\n" % (file_path)
raise ConversionError(str("NO_BLOCK_DATA"))
wide_format_samples = []
for output_samples in wide_format_samples_by_session:
wide_format_samples.extend(output_samples)
scount += len(wide_format_samples)
data_wide = np.array(wide_format_samples, dtype=wide_row_dtype)
data_wide['et_model'] = et_mappings[data_wide['eyetracker_model'][0]]
tracking_eye = parseTrackerMode(data_wide['eyetracker_mode'][0])
if (et_model == 'dpi'):
### Handle DPI data START ###
data_wide, _ = filter_trackloss(filter_bilateral(data_wide), et_model)
# data_wide, _ = filter_trackloss(data_wide, et_model)
if CALIBRATE_DPI:
t1 = getTime()
print "DPI calibration"
if (len(np.unique(data_wide['session_id'])) > 1):
data_wide, _, exp_session_id = handle_dpi_multisession(data_wide)
else:
exp_session_id = data_wide['session_id'][0]
#Calibration settings
poly_type = calibration_settings_set[0]['poly_type']
cal_point_set = calibration_settings_set[0]['cal_point_set']
min_cal_points = calibration_settings_set[0]['min_cal_points']
win_select_func = calibration_settings_set[0]['win_select_func']
units = calibration_settings_set[0]['units']
args={
'win_size': calibration_settings_set[0]['win_size'],
'win_type': calibration_settings_set[0]['win_type'],
'window_skip': calibration_settings_set[0]['window_skip'],
'wsa': [calibration_settings_set[0]['wsa']],
'target_count' : 25
}
cal_points = cal_point_sets[cal_point_set]
cal_block = data_wide['BLOCK'] == 'DPICAL'
exp_block = data_wide['BLOCK'] == 'FS' #TODO: deal with other blocks
DATA_CAL = data_wide[cal_block]
if np.sum(cal_block) == 0:
print 'DPI calibration block missing'
raise ConversionError(str("DPI_CAL_NO_DATA"))
if np.sum(exp_block) == 0:
print 'DPI experiment block missing'
raise ConversionError(str("DPI_EXP_NO_DATA"))
### CALIBRATION START ###
data_wide_raw = np.copy(data_wide)
stim_CAL = win_select_funcs[win_select_func](DATA_CAL, **args)
if len(stim_CAL)>25:
print 'Multiple calibrations per session...skipping'
raise ConversionError(str("DPI_MULTIPLE_CAL_PER_SESS"))
cal_ind=stim_CAL['ROW_INDEX'] == cal_points[:, None]
cal_ind=np.array(np.sum(cal_ind, axis=0), dtype=bool)
### Handle missing calibration points: replace with random ones
stim_key_x = '_'.join((tracking_eye[0], units, 'fix', 'x'))
stim_key_y = '_'.join((tracking_eye[0], units, 'fix', 'y'))
if (np.isnan(stim_CAL[stim_key_x][cal_ind]).any()) | \
(np.isnan(stim_CAL[stim_key_y][cal_ind]).any()):
print 'Calibration points missing...Trying to replace'
valid_cal_ind = np.bitwise_and(np.isfinite(stim_CAL[stim_key_x]),
np.isfinite(stim_CAL[stim_key_y])
)
cal_ind = np.bitwise_and(cal_ind, valid_cal_ind)
extra_cal_ind = np.bitwise_xor(valid_cal_ind, cal_ind)
rand_p_count = cal_point_set-np.sum(cal_ind) #how many points are missing
if rand_p_count > np.sum(extra_cal_ind): #if needed points exceed available points
rand_p_count = np.sum(extra_cal_ind) #select only available amount of extra points
rand_cal_ind = np.random.choice(np.arange(len(extra_cal_ind))[extra_cal_ind], rand_p_count, replace=False)
extra_cal_ind[:]=False
extra_cal_ind[rand_cal_ind] = True
cal_ind = np.bitwise_or(cal_ind, extra_cal_ind)
if (np.sum(cal_ind)<min_cal_points):
print 'Only %d points available..Skipping'%np.sum(cal_ind)
file_log.write('[DPI_CAL_NO_POINTS]\tfile: {file_path}\tAvailable calibration points: {cal_p_available}\n'.format(cal_p_available=np.sum(cal_ind), file_path=file_path ))
raise ConversionError(str("DPI_CAL_NO_POINTS"))
else:
file_log.write('[DPI_CAL_POINTS_REPLACED]\tfile: {file_path}\tReplaced calibration points: {rand_p_count}\n'.format(rand_p_count=rand_p_count, file_path=file_path ))
####
#Conversion to degrees
session_info_dict = getSessionDataFromMsgEvents(hub_file)
session_info = session_info_dict[exp_session_id]
display_size_pix = session_info['display_width_pix'], session_info['display_height_pix']
display_size_mm = screen_measurments['screen_width'], screen_measurments['screen_height']
pix2deg = VisualAngleCalc(display_size_mm, display_size_pix,
screen_measurments['eye_distance']).pix2deg
for eye in tracking_eye:
Px, Py = build_polynomial(stim_CAL['_'.join((eye, units, 'fix_x'))][cal_ind],
stim_CAL['_'.join((eye, units, 'fix_y'))][cal_ind], poly_type)
calX, calY = np.linalg.lstsq(Px, stim_CAL[stim_pos_mappings[units]+'x'][cal_ind])[0], \
np.linalg.lstsq(Py, stim_CAL[stim_pos_mappings[units]+'y'][cal_ind])[0]
Px_data, Py_data = build_polynomial(data_wide_raw['_'.join((eye, units, 'x'))],
data_wide_raw['_'.join((eye, units, 'y'))] , poly_type)
data_wide['_'.join((eye, units, 'x'))] = np.dot(Px_data, calX)
data_wide['_'.join((eye, units, 'y'))] = np.dot(Py_data, calY)
(data_wide['_'.join((eye, 'angle', 'x'))],
data_wide['_'.join((eye, 'angle', 'y'))])=pix2deg(data_wide['_'.join((eye, units, 'x'))],
data_wide['_'.join((eye, units, 'y'))])
### DPI calibration END ###
#Save only FS block
data_wide = data_wide[exp_block]
### Empty recording check
_, loss_count = filter_trackloss(data_wide, et_model)
check_eye = dict()
#Becomes True if all nans found
check_eye['right'] = loss_count['right'] == len(data_wide['right_gaze_x'])
check_eye['left'] = loss_count['left'] == len(data_wide['left_gaze_x'])
if check_eye['right'] & check_eye['left']:
print "Recording does not contain any data...skipping"
raise ConversionError(str("DPI_NO_DATA"))
else:
file_log.write('[DPI_CAL_OK]\tfile: {file_path}\tCalibrated using {cal_p} points\n'.format(cal_p=np.sum(cal_ind), file_path=file_path ))
print 'DPI calibration duration: ', getTime()-t1
### Handle DPI data END ###
else:
### Handle VOG data START ###
data_wide, loss_count = filter_trackloss(data_wide, et_model)
# from edq_shared import filter_bilateral
# data_wide, loss_count = filter_trackloss(filter_bilateral(data_wide), et_model)
# data_wide['et_model'] = data_wide['et_model'][0]+'_0'
#All targets check
if len(np.unique(data_wide['ROW_INDEX']))<25: #continue, if at least half of the targets recorded
print "Not enough data recorded...skipping"
raise ConversionError(str("VOG_NO_ENOUGH_STIM"))
if len(np.unique(data_wide['ROW_INDEX']))!=49:
print "Not all targets recorded"
file_log.write('[VOG_NO_ALL_STIM]\tfile: {file_path}\n'.format(file_path=file_path ))
### Empty recording check
check_eye = dict()
#Becomes True if all nans found
check_eye['right'] = loss_count['right'] == len(data_wide['right_gaze_x'])
check_eye['left'] = loss_count['left'] == len(data_wide['left_gaze_x'])
if check_eye['right'] & check_eye['left']:
print "Recording does not contain any data...skipping"
raise ConversionError(str("VOG_NO_DATA"))
### Data integrity checks
if data_wide['eyetracker_mode'][0] == 'Binocular':
# Tracking mode check
if et_model in MONOCULAR_TRACKERS:
print 'Wrong tracking mode selected for monocular tracker'
raise ConversionError(str("VOG_MODE_SELECT_ERROR"))
# Binocular averaging check
if CHECK_BDA & nan_equal(data_wide['left_gaze_x'], data_wide['right_gaze_x']):
print 'Binocular data averaged..skipping'
raise ConversionError(str("VOG_BDA"))
#Monocular eye select fix
elif check_eye[tracking_eye[0]]:
print "Eye selection error...correcting"
eye_corr = check_eye.keys()[check_eye.values().index(False)].title()
data_wide['eyetracker_mode'] = eye_corr+' eye'
file_log.write('[VOG_EYE_SELECT_CORRECTION]\tfile: {file_path}\n'.format(file_path=file_path ))
###
### Deal with multisession recordings
if (len(np.unique(data_wide['session_id'])) > 1):
print 'Multiple sessions found'
tr_loss = []
session_ids = np.unique(data_wide['session_id'])
### Manual multisession handler
mask_smc = ((special_multisession_cases['subject_id'] == sub )
& (special_multisession_cases['eyetracker_model'] == et_model)
)
if mask_smc.any():
session_ids = special_multisession_cases['session_id'][mask_smc]
print 'Multisession special case'
file_log.write('[VOG_MULTISESSION_SPEC]\tfile: {file_path}\tsid: {sid}\n'.format(file_path=file_path, sid=session_ids[0]))
###
for sid in session_ids:
mask = data_wide['session_id']==sid
if len(np.unique(data_wide['ROW_INDEX'][mask]))==49:
_, loss_count = filter_trackloss(data_wide[mask], et_model)
tr_loss.append((sid, loss_count['avg'], np.sum(mask)))
### PLot multisession data
if plot_multisession:
tr_loss_caption = ''
for eye in parseTrackerMode(data_wide['eyetracker_mode'][0]):
tr_loss_caption += '{eye} eye: {tr_loss:.2f} %; '.format(eye=eye, tr_loss=100*float(loss_count[eye])/np.sum(mask))
_title = 'Tracker: {et_model}, Operator: {operator}, trackloss: {loss}'.format(et_model=et_model,
operator=data_wide['operator'][0],
loss = tr_loss_caption
)
_fname = '{output_dir}/{et_model}/multisession_{et_model}_sub_{sub}_sid_{sid}.png'.format(output_dir=OUTPUT_FOLDER,
et_model=et_model,
sub=sub,
sid=sid
)
plot_data(data_wide, title=_title, fname=_fname, ylim=[-30, 30])
###
tr_loss = np.array(tr_loss)
if tr_loss.any():
least_loss_ind = np.argmin(np.float32(tr_loss[:,1])/tr_loss[:,2])
#Check for 100% trackloss
if tr_loss[least_loss_ind][1] < tr_loss[least_loss_ind][2]:
sid = tr_loss[least_loss_ind,0]
mask = data_wide['session_id']==sid
data_wide = data_wide[mask]
file_log.write('[VOG_MULTISESSION]\tfile: {file_path}\tsid: {sid}\n'.format(file_path=file_path, sid=sid))
else:
print "Session does not contain any data...skipping"
raise ConversionError(str("VOG_NO_DATA_SESSION"))
else:
print "Session does not contain any data...skipping"
file_log.write('[VOG_MULTISESSION]\tfile: {file_path}\tsid: {sid}\n'.format(file_path=file_path, sid=np.nan))
raise ConversionError(str("VOG_NO_DATA_STIM"))
### Handle VOG data END ###
print 'Conversion duration: ', getTime()-t0
data_wide
### Save data
if SAVE_NPY:
np_file_name = r"%s/%s_%s.npy" % (et_dir, et_model, dfile[:-5])
t0 = getTime()
np.save(np_file_name, data_wide)
print 'RAW_NPY save duration: ', getTime()-t0
if SAVE_TXT:
txt_file_name = r"%s/%s_%s.txt" % (et_dir, et_model, dfile[:-5])
t0 = getTime()
save_as_txt(txt_file_name, data_wide)
print 'RAW_TXT save duration: ', getTime()-t0
###
file_log.write('[CONVERSION_OK]\tfile: {file_path}\n'.format(file_path=file_path ))
### Plot data
if et_model in PLOT_TRACKERS:
_, loss_count = filter_trackloss(data_wide, et_model)
tr_loss_caption = ''
for eye in parseTrackerMode(data_wide['eyetracker_mode'][0]):
tr_loss_caption += '{eye} eye: {tr_loss:.2f} %; '.format(eye=eye, tr_loss=100*float(loss_count[eye])/len(data_wide))
_title = 'Tracker: {et_model}, Operator: {operator}, trackloss: {loss}'.format(et_model=et_model,
operator=data_wide['operator'][0],
loss = tr_loss_caption
)
_fname = '{output_dir}/{et_model}/{et_model}_sub_{sub}.png'.format(output_dir=OUTPUT_FOLDER,
et_model=et_model,
sub=sub
)
if et_model == 'dpi':
plot_data(data_wide, title=_title, fname=_fname, ylim=[-10, 10], keep=False)
else:
plot_data(data_wide, title=_title, fname=_fname, ylim=[-30, 30])
###
except ConversionError, e:
print 'Conversion error...skipping'
file_log.write('[{msg}]\tfile: {file_path}\n'.format(msg=e.message, file_path=file_path ))
file_log.write('[CONVERSION_ERROR]\tfile: {file_path}\n'.format(file_path=file_path ))
except:
print 'Unhandled conversion error...skipping'
file_log.write('[UNHANDLED_CONVERSION_ERROR]\tfile: {file_path}\n'.format(file_path=file_path ))
finally:
file_proc_count += 1
print
if hub_file:
hub_file.close()
hub_file = None
if file_log:
file_log.close()
end_time = getTime()
print