To receive notifications about scheduled maintenance, please subscribe to the mailing-list gitlab-operations@sympa.ethz.ch. You can subscribe to the mailing-list at https://sympa.ethz.ch

custom_classes.py 39.3 KB
Newer Older
matthmey's avatar
matthmey committed
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
# from keras.preprocessing.image import *
# This file has been modified from keras (www.keras.io)
from __future__ import absolute_import
from __future__ import print_function

from keras.preprocessing.image import *
to_exclude = ['NumpyArrayIterator']
for name in to_exclude:
    del globals()[name]

import numpy as np
import re
from scipy import linalg
import scipy.ndimage as ndi
from six.moves import range
import os
import threading
import warnings
import multiprocessing.pool
from functools import partial

from keras import backend as K
from keras.utils.data_utils import Sequence
from keras.callbacks import Callback

try:
    from PIL import Image as pil_image
except ImportError:
    pil_image = None

class CustomDataGenerator(object):
    """Generate minibatches of image data with real-time data augmentation.

    # Arguments
        featurewise_center: set input mean to 0 over the dataset.
        samplewise_center: set each sample mean to 0.
        featurewise_std_normalization: divide inputs by std of the dataset.
        samplewise_std_normalization: divide each input by its std.
        zca_whitening: apply ZCA whitening.
        zca_epsilon: epsilon for ZCA whitening. Default is 1e-6.
        rotation_range: degrees (0 to 180).
        width_shift_range: fraction of total width.
        height_shift_range: fraction of total height.
        shear_range: shear intensity (shear angle in radians).
        zoom_range: amount of zoom. if scalar z, zoom will be randomly picked
            in the range [1-z, 1+z]. A sequence of two can be passed instead
            to select this range.
        channel_shift_range: shift range for each channels.
        fill_mode: points outside the boundaries are filled according to the
            given mode ('constant', 'nearest', 'reflect' or 'wrap'). Default
            is 'nearest'.
        cval: value used for points outside the boundaries when fill_mode is
            'constant'. Default is 0.
        horizontal_flip: whether to randomly flip images horizontally.
        vertical_flip: whether to randomly flip images vertically.
        rescale: rescaling factor. If None or 0, no rescaling is applied,
            otherwise we multiply the data by the value provided. This is
            applied after the `preprocessing_function` (if any provided)
            but before any other transformation.
        preprocessing_function: function that will be implied on each input.
            The function will run before any other modification on it.
            The function should take one argument:
            one image (Numpy tensor with rank 3),
            and should output a Numpy tensor with the same shape.
        data_format: 'channels_first' or 'channels_last'. In 'channels_first' mode, the channels dimension
            (the depth) is at index 1, in 'channels_last' mode it is at index 3.
            It defaults to the `image_data_format` value found in your
            Keras config file at `~/.keras/keras.json`.
            If you never set it, then it will be "channels_last".
    """

    def __init__(self,
                 featurewise_center=False,
                 samplewise_center=False,
                 featurewise_std_normalization=False,
                 samplewise_std_normalization=False,
                 zca_whitening=False,
                 zca_epsilon=1e-6,
                 rotation_range=0.,
                 width_shift_range=0.,
                 height_shift_range=0.,
                 shear_range=0.,
                 zoom_range=0.,
                 channel_shift_range=0.,
                 fill_mode='nearest',
                 cval=0.,
                 horizontal_flip=False,
                 vertical_flip=False,
                 rescale=None,
                 preprocessing_function=None,
                 circular_shift = False,
                 crop_spec = False,
                 crop_size = None,
                 scale_range = 0.,
                 add_awgn = False,
                 data_format=None):
        if data_format is None:
            data_format = K.image_data_format()

        self.featurewise_center = featurewise_center
        self.samplewise_center = samplewise_center
        self.featurewise_std_normalization = featurewise_std_normalization
        self.samplewise_std_normalization = samplewise_std_normalization
        self.zca_whitening = zca_whitening
        self.zca_epsilon = zca_epsilon
        self.rotation_range = rotation_range
        self.width_shift_range = width_shift_range
        self.height_shift_range = height_shift_range
        self.shear_range = shear_range
        self.zoom_range = zoom_range
        self.channel_shift_range = channel_shift_range
        self.fill_mode = fill_mode
        self.cval = cval
        self.horizontal_flip = horizontal_flip
        self.vertical_flip = vertical_flip
        self.rescale = rescale
        self.preprocessing_function = preprocessing_function
        self.circular_shift = circular_shift
        self.crop_spec = crop_spec
        self.crop_size = crop_size
        self.add_awgn = add_awgn

        if data_format not in {'channels_last', 'channels_first'}:
            raise ValueError('`data_format` should be `"channels_last"` (channel after row and '
                             'column) or `"channels_first"` (channel before row and column). '
                             'Received arg: ', data_format)
        self.data_format = data_format
        if data_format == 'channels_first':
            self.channel_axis = 1
            self.row_axis = 2
            self.col_axis = 3
        if data_format == 'channels_last':
            self.channel_axis = 3
            self.row_axis = 1
            self.col_axis = 2

        self.mean = None
        self.std = None
        self.principal_components = None

        self.np_iterator = None

        if np.isscalar(zoom_range):
            self.zoom_range = [1 - zoom_range, 1 + zoom_range]
        elif len(zoom_range) == 2:
            self.zoom_range = [zoom_range[0], zoom_range[1]]
        else:
            raise ValueError('`zoom_range` should be a float or '
                             'a tuple or list of two floats. '
                             'Received arg: ', zoom_range)

        if np.isscalar(scale_range):
            self.scale_range = [1 - scale_range, 1 + scale_range]
        elif len(scale_range) == 2:
            self.scale_range = [scale_range[0], scale_range[1]]
        else:
            raise ValueError('`scale_range` should be a float or '
                             'a tuple or list of two floats. '
                             'Received arg: ', scale_range)

    def flow(self, x=None, y=None, use_y=True, dataset=None, batch_size=8, shuffle=True, seed=None,
             save_to_dir=None, save_prefix='', save_format='png',label_whitelist=None, distribution=None):

        # if self.np_iterator is None:
        #     print('init')
        #     self.np_iterator = NumpyArrayIterator(
        #         x, y, self,
        #         batch_size=batch_size,
        #         shuffle=shuffle,
        #         seed=seed,
        #         data_format=self.data_format,
        #         save_to_dir=save_to_dir,
        #         save_prefix=save_prefix,
        #         save_format=save_format)
        #     return self.np_iterator
        # else:
        #     print('here')
        #     self.np_iterator.x = x
        #     self.n = x.shape[0]
        #     if y is not None:
        #         self.np_iterator.y = y
        #     else:
        #         self.np_iterator.y = None
        #     return self.np_iterator
        return NumpyArrayIterator(self,
                x=x, y=y, use_y=use_y,
                dataset = dataset,
                batch_size=batch_size,
                shuffle=shuffle,
                seed=seed,label_whitelist=label_whitelist, distribution=distribution,
                data_format=self.data_format,
                save_to_dir=save_to_dir,
                save_prefix=save_prefix,
                save_format=save_format)

    def flow_from_directory(self, directory,
                            target_size=(256, 256), color_mode='rgb',
                            classes=None, class_mode='categorical',
                            batch_size=8, shuffle=True, seed=None,
                            save_to_dir=None,
                            save_prefix='',
                            save_format='png',
                            follow_links=False):
        return DirectoryIterator(
            directory, self,
            target_size=target_size, color_mode=color_mode,
            classes=classes, class_mode=class_mode,
            data_format=self.data_format,
            batch_size=batch_size, shuffle=shuffle, seed=seed,
            save_to_dir=save_to_dir,
            save_prefix=save_prefix,
            save_format=save_format,
            follow_links=follow_links)

    def standardize(self, x):
        """Apply the normalization configuration to a batch of inputs.

        # Arguments
            x: batch of inputs to be normalized.

        # Returns
            The inputs, normalized.
        """
        if self.preprocessing_function:
            x = self.preprocessing_function(x)
        if self.rescale:
            x *= self.rescale
        # x is a single image, so it doesn't have image number at index 0
        img_channel_axis = self.channel_axis - 1
        if self.samplewise_center:
            x -= np.mean(x, axis=img_channel_axis, keepdims=True)
        if self.samplewise_std_normalization:
            x /= (np.std(x, axis=img_channel_axis, keepdims=True) + 1e-7)

        if self.featurewise_center:
            if self.mean is not None:
                x -= self.mean
            else:
                warnings.warn('This ImageDataGenerator specifies '
                              '`featurewise_center`, but it hasn\'t'
                              'been fit on any training data. Fit it '
                              'first by calling `.fit(numpy_data)`.')
        if self.featurewise_std_normalization:
            if self.std is not None:
                x /= (self.std + 1e-7)
            else:
                warnings.warn('This ImageDataGenerator specifies '
                              '`featurewise_std_normalization`, but it hasn\'t'
                              'been fit on any training data. Fit it '
                              'first by calling `.fit(numpy_data)`.')
        if self.zca_whitening:
            if self.principal_components is not None:
                flatx = np.reshape(x, (-1, np.prod(x.shape[-3:])))
                whitex = np.dot(flatx, self.principal_components)
                x = np.reshape(whitex, x.shape)
            else:
                warnings.warn('This ImageDataGenerator specifies '
                              '`zca_whitening`, but it hasn\'t'
                              'been fit on any training data. Fit it '
                              'first by calling `.fit(numpy_data)`.')
        return x

    def random_transform(self, x, seed=None):
        """Randomly augment a single image tensor.

        # Arguments
            x: 3D tensor, single image.
            seed: random seed.

        # Returns
            A randomly transformed version of the input (same shape).
        """

        # x is a single image, so it doesn't have image number at index 0
        img_row_axis = self.row_axis - 1
        img_col_axis = self.col_axis - 1
        img_channel_axis = self.channel_axis - 1

        if seed is not None:
            np.random.seed(seed)


        if self.add_awgn:
            x = x + np.random.normal(0,np.std(x),x.shape)

        # use composition of homographies
        # to generate final transform that needs to be applied
        if self.rotation_range:
            theta = np.pi / 180 * np.random.uniform(-self.rotation_range, self.rotation_range)
        else:
            theta = 0

        if self.height_shift_range:
            tx = np.random.uniform(-self.height_shift_range, self.height_shift_range) * x.shape[img_row_axis]
        else:
            tx = 0

        if self.width_shift_range:
            ty = np.random.uniform(-self.width_shift_range, self.width_shift_range) * x.shape[img_col_axis]
        else:
            ty = 0

        if self.shear_range:
            shear = np.random.uniform(-self.shear_range, self.shear_range)
        else:
            shear = 0

        if self.zoom_range[0] == 1 and self.zoom_range[1] == 1:
            zx, zy = 1, 1
        else:
            zx, zy = np.random.uniform(self.zoom_range[0], self.zoom_range[1], 2)

        transform_matrix = None
        if theta != 0:
            rotation_matrix = np.array([[np.cos(theta), -np.sin(theta), 0],
                                        [np.sin(theta), np.cos(theta), 0],
                                        [0, 0, 1]])
            transform_matrix = rotation_matrix

        if tx != 0 or ty != 0:
            shift_matrix = np.array([[1, 0, tx],
                                     [0, 1, ty],
                                     [0, 0, 1]])
            transform_matrix = shift_matrix if transform_matrix is None else np.dot(transform_matrix, shift_matrix)

        if shear != 0:
            shear_matrix = np.array([[1, -np.sin(shear), 0],
                                    [0, np.cos(shear), 0],
                                    [0, 0, 1]])
            transform_matrix = shear_matrix if transform_matrix is None else np.dot(transform_matrix, shear_matrix)

        if zx != 1 or zy != 1:
            zoom_matrix = np.array([[zx, 0, 0],
                                    [0, zy, 0],
                                    [0, 0, 1]])
            transform_matrix = zoom_matrix if transform_matrix is None else np.dot(transform_matrix, zoom_matrix)

        if transform_matrix is not None:
            h, w = x.shape[img_row_axis], x.shape[img_col_axis]
            transform_matrix = transform_matrix_offset_center(transform_matrix, h, w)
            x = apply_transform(x, transform_matrix, img_channel_axis,
                                fill_mode=self.fill_mode, cval=self.cval)

        if self.channel_shift_range != 0:
            x = random_channel_shift(x,
                                     self.channel_shift_range,
                                     img_channel_axis)
        if self.horizontal_flip:
            if np.random.random() < 0.5:
                x = flip_axis(x, img_col_axis)

        if self.vertical_flip:
            if np.random.random() < 0.5:
                x = flip_axis(x, img_row_axis)

        if self.circular_shift:
            shift = np.random.randint(x.shape[0])
            x = np.roll(x,shift,0)
            # x = np.roll(x,shift,0)
            # print(x.shape,shift,img_row_axis)
            # off = np.random.randint(x.shape[0]-128)
            # x = x[off:off+128,:,:]

        if self.crop_spec:
            if self.crop_size is None:
                crop_size = x.shape[0]
            else:
                crop_size = self.crop_size
            if x.shape[0] - crop_size > 0:
                off = np.random.randint(x.shape[0]-crop_size)
            else:
                off = 0
            x = x[off:off+crop_size,...]

        if self.scale_range[0] == 1 and self.scale_range[1] == 1:
            scale = 1
        else:
            scale = np.random.uniform(self.scale_range[0], self.scale_range[1], 1)

        x *= scale

        return x

    def fit(self, x,
            augment=False,
            rounds=1,
            seed=None):
        """Fits internal statistics to some sample data.

        Required for featurewise_center, featurewise_std_normalization
        and zca_whitening.

        # Arguments
            x: Numpy array, the data to fit on. Should have rank 4.
                In case of grayscale data,
                the channels axis should have value 1, and in case
                of RGB data, it should have value 3.
            augment: Whether to fit on randomly augmented samples
            rounds: If `augment`,
                how many augmentation passes to do over the data
            seed: random seed.

        # Raises
            ValueError: in case of invalid input `x`.
        """
        x = np.asarray(x, dtype=K.floatx())
        if x.ndim != 4:
            raise ValueError('Input to `.fit()` should have rank 4. '
                             'Got array with shape: ' + str(x.shape))
        if x.shape[self.channel_axis] not in {1, 3, 4}:
            warnings.warn(
                'Expected input to be images (as Numpy array) '
                'following the data format convention "' + self.data_format + '" '
                '(channels on axis ' + str(self.channel_axis) + '), i.e. expected '
                'either 1, 3 or 4 channels on axis ' + str(self.channel_axis) + '. '
                'However, it was passed an array with shape ' + str(x.shape) +
                ' (' + str(x.shape[self.channel_axis]) + ' channels).')

        if seed is not None:
            np.random.seed(seed)

        x = np.copy(x)
        if augment:
            ax = np.zeros(tuple([rounds * x.shape[0]] + list(x.shape)[1:]), dtype=K.floatx())
            for r in range(rounds):
                for i in range(x.shape[0]):
                    ax[i + r * x.shape[0]] = self.random_transform(x[i])
            x = ax

        if self.featurewise_center:
            self.mean = np.mean(x, axis=(0, self.row_axis, self.col_axis))
            broadcast_shape = [1, 1, 1]
            broadcast_shape[self.channel_axis - 1] = x.shape[self.channel_axis]
            self.mean = np.reshape(self.mean, broadcast_shape)
            x -= self.mean

        if self.featurewise_std_normalization:
            self.std = np.std(x, axis=(0, self.row_axis, self.col_axis))
            broadcast_shape = [1, 1, 1]
            broadcast_shape[self.channel_axis - 1] = x.shape[self.channel_axis]
            self.std = np.reshape(self.std, broadcast_shape)
            x /= (self.std + K.epsilon())

        if self.zca_whitening:
            flat_x = np.reshape(x, (x.shape[0], x.shape[1] * x.shape[2] * x.shape[3]))
            sigma = np.dot(flat_x.T, flat_x) / flat_x.shape[0]
            u, s, _ = linalg.svd(sigma)
            self.principal_components = np.dot(np.dot(u, np.diag(1. / np.sqrt(s + self.zca_epsilon))), u.T)




# class Iterator(Sequence):
#     """Base class for image data iterators.

#     Every `Iterator` must implement the `_get_batches_of_transformed_samples`
#     method.

#     # Arguments
#         n: Integer, total number of samples in the dataset to loop over.
#         batch_size: Integer, size of a batch.
#         shuffle: Boolean, whether to shuffle the data between epochs.
#         seed: Random seeding for data shuffling.
#     """

#     def __init__(self, n, batch_size, shuffle, seed):
#         self.n = n
#         self.batch_size = batch_size
#         self.seed = seed
#         self.shuffle = shuffle
#         self.batch_index = 0
#         self.total_batches_seen = 0
#         self.lock = threading.Lock()
#         self.index_array = None
#         self.index_generator = self._flow_index()

#     def _set_index_array(self):
#         self.index_array = np.arange(self.n)
#         if self.shuffle:
#             self.index_array = np.random.permutation(self.n)

#     def __getitem__(self, idx):
#         if idx >= len(self):
#             raise ValueError('Asked to retrieve element {idx}, '
#                              'but the Sequence '
#                              'has length {length}'.format(idx=idx,
#                                                           length=len(self)))
#         if self.seed is not None:
#             np.random.seed(self.seed + self.total_batches_seen)
#         self.total_batches_seen += 1
#         if self.index_array is None:
#             self._set_index_array()
#         index_array = self.index_array[self.batch_size * idx:
#                                        self.batch_size * (idx + 1)]
#         return self._get_batches_of_transformed_samples(index_array)

#     def __len__(self):
#         return (self.n + self.batch_size - 1) // self.batch_size  # round up

#     def on_epoch_end(self):
#         self._set_index_array()

#     def reset(self):
#         self.batch_index = 0

#     def _flow_index(self):
#         # Ensure self.batch_index is 0.
#         self.reset()
#         while 1:
#             if self.seed is not None:
#                 np.random.seed(self.seed + self.total_batches_seen)
#             if self.batch_index == 0:
#                 self._set_index_array()

#             current_index = (self.batch_index * self.batch_size) % self.n
#             if self.n > current_index + self.batch_size:
#                 self.batch_index += 1
#             else:
#                 self.batch_index = 0
#             self.total_batches_seen += 1
#             yield self.index_array[current_index:
#                                    current_index + self.batch_size]

#     def __iter__(self):
#         # Needed if we want to do something like:
#         # for x, y in data_gen.flow(...):
#         return self

#     def __next__(self, *args, **kwargs):
#         return self.next(*args, **kwargs)

#     def _get_batches_of_transformed_samples(self, index_array):
#         """Gets a batch of transformed samples.

#         # Arguments
#             index_array: array of sample indices to include in batch.

#         # Returns
#             A batch of transformed samples.
#         """
#         raise NotImplementedError

class IteratorUniform(Sequence,Callback):
    """Base class for image data iterators.
    Every `Iterator` must implement the `_get_batches_of_transformed_samples`
    method.
    # Arguments
        n: Integer, total number of samples in the dataset to loop over.
        batch_size: Integer, size of a batch.
        shuffle: Boolean, whether to shuffle the data between epochs.
        seed: Random seeding for data shuffling.
    """

    def __init__(self, n, y, batch_size, shuffle, seed, whitelist=None,distribution=None):
        self.batch_size = batch_size
        self.seed = seed
        self.shuffle = shuffle
        self.batch_index = 0
        self.total_batches_seen = 0
        self.lock = threading.Lock()
        self.index_array = None
        self.index_generator = self._flow_index()
        self.epoch_counter = 0
        self.distribution = distribution

        self.n = n

        if self.distribution == 'uniform' or self.distribution == 'one_per_batch':     
            if y is None:
                RuntimeError('If uniform distribution is chosen y must be provided')
            if y.ndim == 1:
                y = y.reshape((-1,1))
            self.num_labels = y.shape[-1]
            if whitelist is None:
                whitelist = np.arange(self.num_labels)
            m = y.nonzero()                                              # this also removes any samples without a label
            self.bin_count = np.array(list(np.bincount(m[1])))
            # label_id = np.argmax(self.bin_count)                            # get the label with the most occurences
            # print(m)
            self.index_list = []
            bin_count_mask = []
            for i in range(self.num_labels):
                # the user can provide a list of labels which should be included when sampling from the dataset
                if i not in whitelist:
                    continue

                index = m[0][m[1] == i]
                if index.size == 0:
                    # print('continue')
                    continue
                self.index_list.append(index)
                bin_count_mask.append(i)

            self.bin_count = self.bin_count[bin_count_mask]         # remove all labels which have not been added to the index_list

            # we would also like to add the samples without an explicit label
            non_label_entries = (np.sum(y,axis=1) == 0).nonzero()
            if len(non_label_entries[0]) > 0:
                self.index_list.append(non_label_entries[0])
                self.bin_count = np.append(self.bin_count,len(non_label_entries[0]))

            # we now must update the num_labels and maxlength
            self.num_labels = len(self.index_list)
            
            # print(self.bin_count)
            self.max_length = np.max(self.bin_count)

            self.fill_threshold = self.n
            self.fill_threshold = np.min(self.bin_count)
            self.fill_threshold_delta = 0

    def _set_index_array(self):
        if self.distribution == 'uniform':
            # self.index_array = np.arange(self.max_length*self.num_labels)
            self.index_array = []
            if self.shuffle:
                for i in range(self.num_labels):
                    np.random.shuffle(self.index_list[i])
            idx = 0

            for j in range(self.max_length):
                for i in range(self.num_labels):
                    idx = j*self.num_labels + i
                    if idx > self.fill_threshold and idx > self.bin_count[i]:
                        continue
                    lidx = j%self.bin_count[i]
                    self.index_array.append(self.index_list[i][lidx].astype(np.int))
                    # print(idx,len(self.index_array),self.n,process.memory_info().rss)

                    # self.index_array[idx] =  self.index_list[i][lidx]
            self.index_array = np.array(self.index_array)
            if self.shuffle:
                np.random.shuffle(self.index_array)

        elif self.distribution == 'one_per_batch':
            # self.index_array = np.arange(self.max_length*self.num_labels)
            self.index_array = np.arange(self.n)
            if self.shuffle:
                np.random.shuffle(self.index_array)
                for i in range(self.num_labels):
                    np.random.shuffle(self.index_list[i])
            idx = 0
            label_idx = 0
            for j in range(0,self.n,self.batch_size):
                for i in range(self.num_labels):
                    label_idx = (label_idx + 1) % self.num_labels
                    if self.fill_threshold >= self.bin_count[label_idx]:
                        lidx = idx%self.bin_count[label_idx]
                        self.index_array[j] = self.index_list[label_idx][lidx].astype(np.int)
                        idx += 1
                        break
            # print('hoi',self.index_array)
            # f=open('tmp/ia.txt','w')
            # for ele in self.index_array:
            #     f.write(str(ele)+'\n')
        else:
            self.index_array = np.arange(self.n)
            if self.shuffle:
                np.random.shuffle(self.index_array)

        # print('\n',len(self.index_array),len(np.unique(self.index_array)),'\n')

        # self.index_array = self.index_array[:len(self.index_array)//self.batch_size*self.batch_size]
        self.n = len(self.index_array)

    def __getitem__(self, idx):
        if idx >= len(self):
            raise ValueError('Asked to retrieve element {idx}, '
                             'but the Sequence '
                             'has length {length}'.format(idx=idx,
                                                          length=len(self)))
        if self.seed is not None:
            np.random.seed(self.seed + self.total_batches_seen)
        self.total_batches_seen += 1
        if self.index_array is None:
            self._set_index_array()
        index_array = self.index_array[self.batch_size * idx:
                                       self.batch_size * (idx + 1)]
        return self._get_batches_of_transformed_samples(index_array)

    def __len__(self):
        return (self.n + self.batch_size - 1) // self.batch_size  # round up

    def on_epoch_end(self,epoch=None,logs=None):
        self.epoch_counter += 1
        if self.distribution == 'uniform':
            self.fill_threshold += self.fill_threshold_delta
            print('\nCurrent fill threshold is',self.fill_threshold,'\n')
        self._set_index_array()

    def reset(self):
        self.batch_index = 0
        self.epoch_counter = 0

    def _flow_index(self):
        # Ensure self.batch_index is 0.
        self.reset()
        while 1:
            if self.seed is not None:
                np.random.seed(self.seed + self.total_batches_seen)
            if self.batch_index == 0:
                self._set_index_array()

            current_index = (self.batch_index * self.batch_size) % self.n
            if self.n > current_index + self.batch_size:
                self.batch_index += 1
            else:
                self.batch_index = 0
            self.total_batches_seen += 1
            yield self.index_array[current_index:
                                   current_index + self.batch_size]

    def __iter__(self):
        # Needed if we want to do something like:
        # for x, y in data_gen.flow(...):
        return self

    def __next__(self, *args, **kwargs):
        return self.next(*args, **kwargs)

    def _get_batches_of_transformed_samples(self, index_array, augment_index_array=None, augment_factor=0.5):
        """Gets a batch of transformed samples.
        # Arguments
            index_array: array of sample indices to include in batch.
        # Returns
            A batch of transformed samples.
        """
        raise NotImplementedError


class NumpyArrayIterator(IteratorUniform):
    """Iterator yielding data from a Numpy array.

    # Arguments
        x: Numpy array of input data.
        y: Numpy array of targets data.
        image_data_generator: Instance of `ImageDataGenerator`
            to use for random transformations and normalization.
        batch_size: Integer, size of a batch.
        shuffle: Boolean, whether to shuffle the data between epochs.
        seed: Random seed for data shuffling.
        data_format: String, one of `channels_first`, `channels_last`.
        save_to_dir: Optional directory where to save the pictures
            being yielded, in a viewable format. This is useful
            for visualizing the random transformations being
            applied, for debugging purposes.
        save_prefix: String prefix to use for saving sample
            images (if `save_to_dir` is set).
        save_format: Format to use for saving sample images
            (if `save_to_dir` is set).
    """

    def __init__(self, image_data_generator, x=None, y=None, use_y=True, dataset=None,
                 batch_size=8, shuffle=False, seed=None,
                 data_format=None, label_whitelist = None, distribution=None,
                 save_to_dir=None, save_prefix='', save_format='png'):
        # if y is not None and len(x) != len(y):
        #     raise ValueError('X (images tensor) and y (labels) '
        #                      'should have the same length. '
        #                      'Found: X.shape = %s, y.shape = %s' %
        #                      (np.asarray(x).shape, np.asarray(y).shape))

        if x is None and dataset is None:
            RuntimeError('At least either x or a dataset must be provided')

        if data_format is None:
            data_format = K.image_data_format()

        if x is not None:
            self.x = np.asarray(x, dtype=K.floatx())
            # if self.x.ndim != 4:
            #     raise ValueError('Input data in `NumpyArrayIterator` '
            #                     'should have rank 4. You passed an array '
            #                     'with shape', self.x.shape)
            channels_axis = -1 if data_format == 'channels_last' else 1
            if self.x.shape[channels_axis] not in {1, 3, 4}:
                warnings.warn('NumpyArrayIterator is set to use the '
                            'data format convention "' + data_format + '" '
                            '(channels on axis ' + str(channels_axis) + '), i.e. expected '
                            'either 1, 3 or 4 channels on axis ' + str(channels_axis) + '. '
                            'However, it was passed an array with shape ' + str(self.x.shape) +
                            ' (' + str(self.x.shape[channels_axis]) + ' channels).')

        self.use_y = use_y
        if y is not None:
            self.y = np.asarray(y)
            self.y_shape = self.y.shape
        else:
            self.y = None
            self.y_shape = None

        self.image_data_generator = image_data_generator
        self.data_format = data_format
        self.save_to_dir = save_to_dir
        self.save_prefix = save_prefix
        self.save_format = save_format

        self.dataset = dataset

        if dataset is None:
            x_example = self.x[0]
            num_samples = self.x.shape[0]
        else:
            x,y = self.dataset.get_segment(0,1)
            x_example = x[0]
            num_samples = self.dataset.shape[0]

            if self.y is None and self.use_y:
                self.y = self.dataset.get_y()
                self.y_shape = tuple([num_samples] + list(y[0].shape))
            if self.y is not None:
                if self.y.shape[0] != num_samples:
                    raise RuntimeError('When using a dataset with a different y, please make sure that the number of samples match')

        x_example = self.image_data_generator.random_transform(x_example.astype(K.floatx()))
        x_example = self.image_data_generator.standardize(x_example)
        
        self.shape = tuple([num_samples] + list(x_example.shape))
        
        if self.y is None and distribution == 'uniform':
            RuntimeError('If uniform distribution is chosen y must be provided')
                    
        self.label_whitelist_idx = None
        if label_whitelist is not None:
            if isinstance(label_whitelist[0],str):
                self.label_whitelist_idx = []
                for label in label_whitelist:
                    self.label_whitelist_idx.append(dataset.get_label(label))
            elif isinstance(label_whitelist[0],int):
                self.label_whitelist_idx = label_whitelist
            else:
                raise ValueError('Please provide a label_whitelist as either integer indices or list of strings. For list of strings a dataset must be provided via the dataset argument')
            
            if self.y is not None:
                self.y_shape = (num_samples,len(label_whitelist))
        

        super(NumpyArrayIterator, self).__init__(self.shape[0], self.y, batch_size, shuffle, seed, whitelist=self.label_whitelist_idx, distribution=distribution)

    def _get_batches_of_transformed_samples(self, index_array, augment_index_array=None, augment_factor=0.5):
        # print(index_array)            
        batch_x = np.zeros(tuple([len(index_array)] + list(self.shape)[1:]), dtype=K.floatx())

        batch_y = None
        if self.use_y:
            batch_y = np.zeros(tuple([len(index_array)] + list(self.y_shape)[1:]), dtype=K.floatx())
        
        for i, j in enumerate(index_array):
            if self.dataset is None:
                x = self.x[j]
                if augment_index_array is not None:
                    x = x + self.x[augment_index_array[i]] * augment_factor
            else:
                x,y = self.dataset.get_segment(j,1)
                x = x[0]
                
                if augment_index_array is not None:
                    x_aug,y_aug = self.dataset.get_segment(augment_index_array[i],1)
                    x = x + x_aug[0] * augment_factor

                if self.use_y:
                    y = self.y[j]

                    if self.label_whitelist_idx is None:
                        batch_y[i] = y
                    else:
                        batch_y[i] = y[self.label_whitelist_idx]


                

            x = self.image_data_generator.random_transform(x.astype(K.floatx()))
            x = self.image_data_generator.standardize(x)
            batch_x[i] = x
        if self.save_to_dir:
            for i, j in enumerate(index_array):
                img = array_to_img(batch_x[i], self.data_format, scale=True)
                fname = '{prefix}_{index}_{hash}.{format}'.format(prefix=self.save_prefix,
                                                                  index=j,
                                                                  hash=np.random.randint(1e4),
                                                                  format=self.save_format)
                img.save(os.path.join(self.save_to_dir, fname))

        if self.dataset is None and self.y is not None:
            if self.label_whitelist_idx is None:
                batch_y = self.y[index_array]
            else:
                batch_y = self.y[index_array,self.label_whitelist_idx]

        return batch_x, batch_y

    def next(self):
        """For python 2.x.

        # Returns
            The next batch.
        """
        # Keeps under lock only the mechanism which advances
        # the indexing of each batch.
        with self.lock:
            index_array = next(self.index_generator)

        # print(index_array)
        # The transformation of images is not under thread lock
        # so it can be done in parallel
        return self._get_batches_of_transformed_samples(index_array)


# def _count_valid_files_in_directory(directory, white_list_formats, follow_links):
#     """Count files with extension in `white_list_formats` contained in a directory.

#     # Arguments
#         directory: absolute path to the directory containing files to be counted
#         white_list_formats: set of strings containing allowed extensions for
#             the files to be counted.

#     # Returns
#         the count of files with extension in `white_list_formats` contained in
#         the directory.
#     """
#     def _recursive_list(subpath):
#         return sorted(os.walk(subpath, followlinks=follow_links), key=lambda tpl: tpl[0])

#     samples = 0
#     for root, _, files in _recursive_list(directory):
#         for fname in files:
#             is_valid = False
#             for extension in white_list_formats:
#                 if fname.lower().endswith('.' + extension):
#                     is_valid = True
#                     break
#             if is_valid:
#                 samples += 1
#     return samples


# def _list_valid_filenames_in_directory(directory, white_list_formats,
#                                        class_indices, follow_links):
#     """List paths of files in `subdir` relative from `directory` whose extensions are in `white_list_formats`.

#     # Arguments
#         directory: absolute path to a directory containing the files to list.
#             The directory name is used as class label and must be a key of `class_indices`.
#         white_list_formats: set of strings containing allowed extensions for
#             the files to be counted.
#         class_indices: dictionary mapping a class name to its index.

#     # Returns
#         classes: a list of class indices
#         filenames: the path of valid files in `directory`, relative from
#             `directory`'s parent (e.g., if `directory` is "dataset/class1",
#             the filenames will be ["class1/file1.jpg", "class1/file2.jpg", ...]).
#     """
#     def _recursive_list(subpath):
#         return sorted(os.walk(subpath, followlinks=follow_links), key=lambda tpl: tpl[0])

#     classes = []
#     filenames = []
#     subdir = os.path.basename(directory)
#     basedir = os.path.dirname(directory)
#     for root, _, files in _recursive_list(directory):
#         for fname in sorted(files):
#             is_valid = False
#             for extension in white_list_formats:
#                 if fname.lower().endswith('.' + extension):
#                     is_valid = True
#                     break
#             if is_valid:
#                 classes.append(class_indices[subdir])
#                 # add filename relative to directory
#                 absolute_path = os.path.join(root, fname)
#                 filenames.append(os.path.relpath(absolute_path, basedir))
#     return classes, filenames