test_transforms.py 34.6 KB
Newer Older
1
2
import torch
import torchvision.transforms as transforms
3
import torchvision.transforms.functional as F
4
5
import unittest
import random
6
import numpy as np
7
8
9
10
11
12
from PIL import Image
try:
    import accimage
except ImportError:
    accimage = None

13
14
15
16
17
try:
    from scipy import stats
except ImportError:
    stats = None

18
GRACE_HOPPER = 'assets/grace_hopper_517x606.jpg'
19

20

21
class Tester(unittest.TestCase):
22

23
24
25
26
    def test_crop(self):
        height = random.randint(10, 32) * 2
        width = random.randint(10, 32) * 2
        oheight = random.randint(5, (height - 2) / 2) * 2
27
28
        owidth = random.randint(5, (width - 2) / 2) * 2

29
        img = torch.ones(3, height, width)
30
31
32
        oh1 = (height - oheight) // 2
        ow1 = (width - owidth) // 2
        imgnarrow = img[:, oh1:oh1 + oheight, ow1:ow1 + owidth]
33
34
35
36
37
38
39
        imgnarrow.fill_(0)
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.CenterCrop((oheight, owidth)),
            transforms.ToTensor(),
        ])(img)
        assert result.sum() == 0, "height: " + str(height) + " width: " \
40
                                  + str(width) + " oheight: " + str(oheight) + " owidth: " + str(owidth)
41
42
43
44
45
46
47
48
49
        oheight += 1
        owidth += 1
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.CenterCrop((oheight, owidth)),
            transforms.ToTensor(),
        ])(img)
        sum1 = result.sum()
        assert sum1 > 1, "height: " + str(height) + " width: " \
50
                         + str(width) + " oheight: " + str(oheight) + " owidth: " + str(owidth)
51
        oheight += 1
52
        owidth += 1
53
54
55
56
57
58
59
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.CenterCrop((oheight, owidth)),
            transforms.ToTensor(),
        ])(img)
        sum2 = result.sum()
        assert sum2 > 0, "height: " + str(height) + " width: " \
60
                         + str(width) + " oheight: " + str(oheight) + " owidth: " + str(owidth)
61
        assert sum2 > sum1, "height: " + str(height) + " width: " \
62
                            + str(width) + " oheight: " + str(oheight) + " owidth: " + str(owidth)
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
    def test_five_crop(self):
        to_pil_image = transforms.ToPILImage()
        h = random.randint(5, 25)
        w = random.randint(5, 25)
        for single_dim in [True, False]:
            crop_h = random.randint(1, h)
            crop_w = random.randint(1, w)
            if single_dim:
                crop_h = min(crop_h, crop_w)
                crop_w = crop_h
                transform = transforms.FiveCrop(crop_h)
            else:
                transform = transforms.FiveCrop((crop_h, crop_w))

            img = torch.FloatTensor(3, h, w).uniform_()
            results = transform(to_pil_image(img))

            assert len(results) == 5
            for crop in results:
                assert crop.size == (crop_w, crop_h)

            to_pil_image = transforms.ToPILImage()
            tl = to_pil_image(img[:, 0:crop_h, 0:crop_w])
            tr = to_pil_image(img[:, 0:crop_h, w - crop_w:])
            bl = to_pil_image(img[:, h - crop_h:, 0:crop_w])
            br = to_pil_image(img[:, h - crop_h:, w - crop_w:])
            center = transforms.CenterCrop((crop_h, crop_w))(to_pil_image(img))
            expected_output = (tl, tr, bl, br, center)
            assert results == expected_output

    def test_ten_crop(self):
        to_pil_image = transforms.ToPILImage()
        h = random.randint(5, 25)
        w = random.randint(5, 25)
        for should_vflip in [True, False]:
            for single_dim in [True, False]:
                crop_h = random.randint(1, h)
                crop_w = random.randint(1, w)
                if single_dim:
                    crop_h = min(crop_h, crop_w)
                    crop_w = crop_h
105
106
                    transform = transforms.TenCrop(crop_h,
                                                   vertical_flip=should_vflip)
107
108
                    five_crop = transforms.FiveCrop(crop_h)
                else:
109
110
                    transform = transforms.TenCrop((crop_h, crop_w),
                                                   vertical_flip=should_vflip)
111
112
113
114
115
                    five_crop = transforms.FiveCrop((crop_h, crop_w))

                img = to_pil_image(torch.FloatTensor(3, h, w).uniform_())
                results = transform(img)
                expected_output = five_crop(img)
116
117
118
119
120

                # Checking if FiveCrop and TenCrop can be printed as string
                transform.__repr__()
                five_crop.__repr__()

121
122
123
124
125
126
127
128
129
130
                if should_vflip:
                    vflipped_img = img.transpose(Image.FLIP_TOP_BOTTOM)
                    expected_output += five_crop(vflipped_img)
                else:
                    hflipped_img = img.transpose(Image.FLIP_LEFT_RIGHT)
                    expected_output += five_crop(hflipped_img)

                assert len(results) == 10
                assert expected_output == results

131
    def test_resize(self):
132
133
134
        height = random.randint(24, 32) * 2
        width = random.randint(24, 32) * 2
        osize = random.randint(5, 12) * 2
135

136
137
138
        img = torch.ones(3, height, width)
        result = transforms.Compose([
            transforms.ToPILImage(),
139
            transforms.Resize(osize),
140
141
142
143
            transforms.ToTensor(),
        ])(img)
        assert osize in result.size()
        if height < width:
144
            assert result.size(1) <= result.size(2)
145
146
147
        elif width < height:
            assert result.size(1) >= result.size(2)

148
149
        result = transforms.Compose([
            transforms.ToPILImage(),
150
            transforms.Resize([osize, osize]),
151
152
153
154
155
156
            transforms.ToTensor(),
        ])(img)
        assert osize in result.size()
        assert result.size(1) == osize
        assert result.size(2) == osize

157
158
159
160
        oheight = random.randint(5, 12) * 2
        owidth = random.randint(5, 12) * 2
        result = transforms.Compose([
            transforms.ToPILImage(),
161
            transforms.Resize((oheight, owidth)),
162
163
164
165
166
167
168
            transforms.ToTensor(),
        ])(img)
        assert result.size(1) == oheight
        assert result.size(2) == owidth

        result = transforms.Compose([
            transforms.ToPILImage(),
169
            transforms.Resize([oheight, owidth]),
170
171
172
173
174
            transforms.ToTensor(),
        ])(img)
        assert result.size(1) == oheight
        assert result.size(2) == owidth

175
176
177
178
    def test_random_crop(self):
        height = random.randint(10, 32) * 2
        width = random.randint(10, 32) * 2
        oheight = random.randint(5, (height - 2) / 2) * 2
179
        owidth = random.randint(5, (width - 2) / 2) * 2
180
181
182
183
184
185
186
187
188
        img = torch.ones(3, height, width)
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.RandomCrop((oheight, owidth)),
            transforms.ToTensor(),
        ])(img)
        assert result.size(1) == oheight
        assert result.size(2) == owidth

189
190
191
192
193
194
195
196
197
        padding = random.randint(1, 20)
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.RandomCrop((oheight, owidth), padding=padding),
            transforms.ToTensor(),
        ])(img)
        assert result.size(1) == oheight
        assert result.size(2) == owidth

198
199
200
201
202
203
204
205
206
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.RandomCrop((height, width)),
            transforms.ToTensor()
        ])(img)
        assert result.size(1) == height
        assert result.size(2) == width
        assert np.allclose(img.numpy(), result.numpy())

207
208
209
210
211
212
213
214
215
216
    def test_pad(self):
        height = random.randint(10, 32) * 2
        width = random.randint(10, 32) * 2
        img = torch.ones(3, height, width)
        padding = random.randint(1, 20)
        result = transforms.Compose([
            transforms.ToPILImage(),
            transforms.Pad(padding),
            transforms.ToTensor(),
        ])(img)
217
218
        assert result.size(1) == height + 2 * padding
        assert result.size(2) == width + 2 * padding
Soumith Chintala's avatar
Soumith Chintala committed
219

220
221
222
223
224
225
226
227
228
229
230
231
232
233
    def test_pad_with_tuple_of_pad_values(self):
        height = random.randint(10, 32) * 2
        width = random.randint(10, 32) * 2
        img = transforms.ToPILImage()(torch.ones(3, height, width))

        padding = tuple([random.randint(1, 20) for _ in range(2)])
        output = transforms.Pad(padding)(img)
        assert output.size == (width + padding[0] * 2, height + padding[1] * 2)

        padding = tuple([random.randint(1, 20) for _ in range(4)])
        output = transforms.Pad(padding)(img)
        assert output.size[0] == width + padding[0] + padding[2]
        assert output.size[1] == height + padding[1] + padding[3]

234
235
236
        # Checking if Padding can be printed as string
        transforms.Pad(padding).__repr__()

237
    def test_pad_raises_with_invalid_pad_sequence_len(self):
238
239
240
241
242
243
244
245
246
        with self.assertRaises(ValueError):
            transforms.Pad(())

        with self.assertRaises(ValueError):
            transforms.Pad((1, 2, 3))

        with self.assertRaises(ValueError):
            transforms.Pad((1, 2, 3, 4, 5))

Soumith Chintala's avatar
Soumith Chintala committed
247
248
249
250
    def test_lambda(self):
        trans = transforms.Lambda(lambda x: x.add(10))
        x = torch.randn(10)
        y = trans(x)
251
        assert (y.equal(torch.add(x, 10)))
Soumith Chintala's avatar
Soumith Chintala committed
252
253
254
255

        trans = transforms.Lambda(lambda x: x.add_(10))
        x = torch.randn(10)
        y = trans(x)
256
257
        assert (y.equal(x))

258
259
260
        # Checking if Lambda can be printed as string
        trans.__repr__()

261
    def test_to_tensor(self):
262
        test_channels = [1, 3, 4]
263
264
        height, width = 4, 4
        trans = transforms.ToTensor()
265
266
267
268
269
270
271
272
273
274
275

        for channels in test_channels:
            input_data = torch.ByteTensor(channels, height, width).random_(0, 255).float().div_(255)
            img = transforms.ToPILImage()(input_data)
            output = trans(img)
            assert np.allclose(input_data.numpy(), output.numpy())

            ndarray = np.random.randint(low=0, high=255, size=(height, width, channels))
            output = trans(ndarray)
            expected_output = ndarray.transpose((2, 0, 1)) / 255.0
            assert np.allclose(output.numpy(), expected_output)
276

277
278
279
280
281
282
283
284
285
286
287
288
289
    @unittest.skipIf(accimage is None, 'accimage not available')
    def test_accimage_to_tensor(self):
        trans = transforms.ToTensor()

        expected_output = trans(Image.open(GRACE_HOPPER).convert('RGB'))
        output = trans(accimage.Image(GRACE_HOPPER))

        self.assertEqual(expected_output.size(), output.size())
        assert np.allclose(output.numpy(), expected_output.numpy())

    @unittest.skipIf(accimage is None, 'accimage not available')
    def test_accimage_resize(self):
        trans = transforms.Compose([
290
            transforms.Resize(256, interpolation=Image.LINEAR),
291
292
293
            transforms.ToTensor(),
        ])

294
295
296
        # Checking if Compose, Resize and ToTensor can be printed as string
        trans.__repr__()

297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
        expected_output = trans(Image.open(GRACE_HOPPER).convert('RGB'))
        output = trans(accimage.Image(GRACE_HOPPER))

        self.assertEqual(expected_output.size(), output.size())
        self.assertLess(np.abs((expected_output - output).mean()), 1e-3)
        self.assertLess((expected_output - output).var(), 1e-5)
        # note the high absolute tolerance
        assert np.allclose(output.numpy(), expected_output.numpy(), atol=5e-2)

    @unittest.skipIf(accimage is None, 'accimage not available')
    def test_accimage_crop(self):
        trans = transforms.Compose([
            transforms.CenterCrop(256),
            transforms.ToTensor(),
        ])

313
314
315
        # Checking if Compose, CenterCrop and ToTensor can be printed as string
        trans.__repr__()

316
317
318
319
320
321
        expected_output = trans(Image.open(GRACE_HOPPER).convert('RGB'))
        output = trans(accimage.Image(GRACE_HOPPER))

        self.assertEqual(expected_output.size(), output.size())
        assert np.allclose(output.numpy(), expected_output.numpy())

322
    def test_1_channel_tensor_to_pil_image(self):
323
324
        to_tensor = transforms.ToTensor()

325
        img_data_float = torch.Tensor(1, 4, 4).uniform_()
326
327
328
329
        img_data_byte = torch.ByteTensor(1, 4, 4).random_(0, 255)
        img_data_short = torch.ShortTensor(1, 4, 4).random_()
        img_data_int = torch.IntTensor(1, 4, 4).random_()

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
        inputs = [img_data_float, img_data_byte, img_data_short, img_data_int]
        expected_outputs = [img_data_float.mul(255).int().float().div(255).numpy(),
                            img_data_byte.float().div(255.0).numpy(),
                            img_data_short.numpy(),
                            img_data_int.numpy()]
        expected_modes = ['L', 'L', 'I;16', 'I']

        for img_data, expected_output, mode in zip(inputs, expected_outputs, expected_modes):
            for transform in [transforms.ToPILImage(), transforms.ToPILImage(mode=mode)]:
                img = transform(img_data)
                assert img.mode == mode
                assert np.allclose(expected_output, to_tensor(img).numpy())

    def test_1_channel_ndarray_to_pil_image(self):
        img_data_float = torch.Tensor(4, 4, 1).uniform_().numpy()
        img_data_byte = torch.ByteTensor(4, 4, 1).random_(0, 255).numpy()
        img_data_short = torch.ShortTensor(4, 4, 1).random_().numpy()
        img_data_int = torch.IntTensor(4, 4, 1).random_().numpy()

        inputs = [img_data_float, img_data_byte, img_data_short, img_data_int]
        expected_modes = ['F', 'L', 'I;16', 'I']
        for img_data, mode in zip(inputs, expected_modes):
            for transform in [transforms.ToPILImage(), transforms.ToPILImage(mode=mode)]:
                img = transform(img_data)
                assert img.mode == mode
                assert np.allclose(img_data[:, :, 0], img)

    def test_3_channel_tensor_to_pil_image(self):
        def verify_img_data(img_data, expected_output, mode):
            if mode is None:
                img = transforms.ToPILImage()(img_data)
                assert img.mode == 'RGB'  # default should assume RGB
            else:
                img = transforms.ToPILImage(mode=mode)(img_data)
                assert img.mode == mode
            split = img.split()
            for i in range(3):
367
                assert np.allclose(expected_output[i].numpy(), F.to_tensor(split[i]).numpy())
368

369
370
371
372
        img_data = torch.Tensor(3, 4, 4).uniform_()
        expected_output = img_data.mul(255).int().float().div(255)
        for mode in [None, 'RGB', 'HSV', 'YCbCr']:
            verify_img_data(img_data, expected_output, mode=mode)
373

374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
        with self.assertRaises(ValueError):
            # should raise if we try a mode for 4 or 1 channel images
            transforms.ToPILImage(mode='RGBA')(img_data)
            transforms.ToPILImage(mode='P')(img_data)

    def test_3_channel_ndarray_to_pil_image(self):
        def verify_img_data(img_data, mode):
            if mode is None:
                img = transforms.ToPILImage()(img_data)
                assert img.mode == 'RGB'  # default should assume RGB
            else:
                img = transforms.ToPILImage(mode=mode)(img_data)
                assert img.mode == mode
            split = img.split()
            for i in range(3):
                assert np.allclose(img_data[:, :, i], split[i])
390

391
392
393
394
        img_data = torch.ByteTensor(4, 4, 3).random_(0, 255).numpy()
        for mode in [None, 'RGB', 'HSV', 'YCbCr']:
            verify_img_data(img_data, mode)

395
396
397
        # Checking if ToPILImage can be printed as string
        transforms.ToPILImage().__repr__()

398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
        with self.assertRaises(ValueError):
            # should raise if we try a mode for 4 or 1 channel images
            transforms.ToPILImage(mode='RGBA')(img_data)
            transforms.ToPILImage(mode='P')(img_data)

    def test_4_channel_tensor_to_pil_image(self):
        def verify_img_data(img_data, expected_output, mode):
            if mode is None:
                img = transforms.ToPILImage()(img_data)
                assert img.mode == 'RGBA'  # default should assume RGBA
            else:
                img = transforms.ToPILImage(mode=mode)(img_data)
                assert img.mode == mode

            split = img.split()
            for i in range(4):
414
                assert np.allclose(expected_output[i].numpy(), F.to_tensor(split[i]).numpy())
415

416
        img_data = torch.Tensor(4, 4, 4).uniform_()
417
        expected_output = img_data.mul(255).int().float().div(255)
418
419
        for mode in [None, 'RGBA', 'CMYK']:
            verify_img_data(img_data, expected_output, mode)
420

421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
        with self.assertRaises(ValueError):
            # should raise if we try a mode for 3 or 1 channel images
            transforms.ToPILImage(mode='RGB')(img_data)
            transforms.ToPILImage(mode='P')(img_data)

    def test_4_channel_ndarray_to_pil_image(self):
        def verify_img_data(img_data, mode):
            if mode is None:
                img = transforms.ToPILImage()(img_data)
                assert img.mode == 'RGBA'  # default should assume RGBA
            else:
                img = transforms.ToPILImage(mode=mode)(img_data)
                assert img.mode == mode
            split = img.split()
            for i in range(4):
                assert np.allclose(img_data[:, :, i], split[i])
437

438
439
440
        img_data = torch.ByteTensor(4, 4, 4).random_(0, 255).numpy()
        for mode in [None, 'RGBA', 'CMYK']:
            verify_img_data(img_data, mode)
441

442
443
444
445
        with self.assertRaises(ValueError):
            # should raise if we try a mode for 3 or 1 channel images
            transforms.ToPILImage(mode='RGB')(img_data)
            transforms.ToPILImage(mode='P')(img_data)
446

447
    def test_ndarray_bad_types_to_pil_image(self):
448
        trans = transforms.ToPILImage()
449
        with self.assertRaises(TypeError):
450
451
452
453
454
            trans(np.ones([4, 4, 1], np.int64))
            trans(np.ones([4, 4, 1], np.uint16))
            trans(np.ones([4, 4, 1], np.uint32))
            trans(np.ones([4, 4, 1], np.float64))

455
456
    @unittest.skipIf(stats is None, 'scipy.stats not available')
    def test_random_vertical_flip(self):
457
458
        random_state = random.getstate()
        random.seed(42)
459
460
461
        img = transforms.ToPILImage()(torch.rand(3, 10, 10))
        vimg = img.transpose(Image.FLIP_TOP_BOTTOM)

462
        num_samples = 250
463
        num_vertical = 0
464
        for _ in range(num_samples):
465
466
467
468
            out = transforms.RandomVerticalFlip()(img)
            if out == vimg:
                num_vertical += 1

469
470
471
        p_value = stats.binom_test(num_vertical, num_samples, p=0.5)
        random.setstate(random_state)
        assert p_value > 0.0001
472

473
474
475
        # Checking if RandomVerticalFlip can be printed as string
        transforms.RandomVerticalFlip().__repr__()

476
477
    @unittest.skipIf(stats is None, 'scipy.stats not available')
    def test_random_horizontal_flip(self):
478
479
        random_state = random.getstate()
        random.seed(42)
480
481
482
        img = transforms.ToPILImage()(torch.rand(3, 10, 10))
        himg = img.transpose(Image.FLIP_LEFT_RIGHT)

483
        num_samples = 250
484
        num_horizontal = 0
485
        for _ in range(num_samples):
486
487
488
489
            out = transforms.RandomHorizontalFlip()(img)
            if out == himg:
                num_horizontal += 1

490
491
492
        p_value = stats.binom_test(num_horizontal, num_samples, p=0.5)
        random.setstate(random_state)
        assert p_value > 0.0001
493

494
495
496
        # Checking if RandomHorizontalFlip can be printed as string
        transforms.RandomHorizontalFlip().__repr__()

497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
    @unittest.skipIf(stats is None, 'scipt.stats is not available')
    def test_normalize(self):
        def samples_from_standard_normal(tensor):
            p_value = stats.kstest(list(tensor.view(-1)), 'norm', args=(0, 1)).pvalue
            return p_value > 0.0001

        random_state = random.getstate()
        random.seed(42)
        for channels in [1, 3]:
            img = torch.rand(channels, 10, 10)
            mean = [img[c].mean() for c in range(channels)]
            std = [img[c].std() for c in range(channels)]
            normalized = transforms.Normalize(mean, std)(img)
            assert samples_from_standard_normal(normalized)
        random.setstate(random_state)

513
514
515
        # Checking if Normalize can be printed as string
        transforms.Normalize(mean, std).__repr__()

516
517
518
519
520
521
522
    def test_adjust_brightness(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')

        # test 0
523
        y_pil = F.adjust_brightness(x_pil, 1)
524
525
526
527
        y_np = np.array(y_pil)
        assert np.allclose(y_np, x_np)

        # test 1
528
        y_pil = F.adjust_brightness(x_pil, 0.5)
529
530
531
532
533
534
        y_np = np.array(y_pil)
        y_ans = [0, 2, 6, 27, 67, 113, 18, 4, 117, 45, 127, 0]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 2
535
        y_pil = F.adjust_brightness(x_pil, 2)
536
537
538
539
540
541
542
543
544
545
546
547
        y_np = np.array(y_pil)
        y_ans = [0, 10, 26, 108, 255, 255, 74, 16, 255, 180, 255, 2]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

    def test_adjust_contrast(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')

        # test 0
548
        y_pil = F.adjust_contrast(x_pil, 1)
549
550
551
552
        y_np = np.array(y_pil)
        assert np.allclose(y_np, x_np)

        # test 1
553
        y_pil = F.adjust_contrast(x_pil, 0.5)
554
555
556
557
558
559
        y_np = np.array(y_pil)
        y_ans = [43, 45, 49, 70, 110, 156, 61, 47, 160, 88, 170, 43]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 2
560
        y_pil = F.adjust_contrast(x_pil, 2)
561
562
563
564
565
566
567
568
569
570
571
572
        y_np = np.array(y_pil)
        y_ans = [0, 0, 0, 22, 184, 255, 0, 0, 255, 94, 255, 0]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

    def test_adjust_saturation(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')

        # test 0
573
        y_pil = F.adjust_saturation(x_pil, 1)
574
575
576
577
        y_np = np.array(y_pil)
        assert np.allclose(y_np, x_np)

        # test 1
578
        y_pil = F.adjust_saturation(x_pil, 0.5)
579
580
581
582
583
584
        y_np = np.array(y_pil)
        y_ans = [2, 4, 8, 87, 128, 173, 39, 25, 138, 133, 215, 88]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 2
585
        y_pil = F.adjust_saturation(x_pil, 2)
586
587
588
589
590
591
592
593
594
595
596
597
        y_np = np.array(y_pil)
        y_ans = [0, 6, 22, 0, 149, 255, 32, 0, 255, 4, 255, 0]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

    def test_adjust_hue(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')

        with self.assertRaises(ValueError):
598
599
            F.adjust_hue(x_pil, -0.7)
            F.adjust_hue(x_pil, 1)
600
601
602

        # test 0: almost same as x_data but not exact.
        # probably because hsv <-> rgb floating point ops
603
        y_pil = F.adjust_hue(x_pil, 0)
604
605
606
607
608
609
        y_np = np.array(y_pil)
        y_ans = [0, 5, 13, 54, 139, 226, 35, 8, 234, 91, 255, 1]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 1
610
        y_pil = F.adjust_hue(x_pil, 0.25)
611
612
613
614
615
616
        y_np = np.array(y_pil)
        y_ans = [13, 0, 12, 224, 54, 226, 234, 8, 99, 1, 222, 255]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 2
617
        y_pil = F.adjust_hue(x_pil, -0.25)
618
619
620
621
622
623
624
625
626
627
628
629
        y_np = np.array(y_pil)
        y_ans = [0, 13, 2, 54, 226, 58, 8, 234, 152, 255, 43, 1]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

    def test_adjust_gamma(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')

        # test 0
630
        y_pil = F.adjust_gamma(x_pil, 1)
631
632
633
634
        y_np = np.array(y_pil)
        assert np.allclose(y_np, x_np)

        # test 1
635
        y_pil = F.adjust_gamma(x_pil, 0.5)
636
637
638
639
640
641
        y_np = np.array(y_pil)
        y_ans = [0, 35, 57, 117, 185, 240, 97, 45, 244, 151, 255, 15]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

        # test 2
642
        y_pil = F.adjust_gamma(x_pil, 2)
643
644
645
646
647
648
649
650
651
652
653
654
        y_np = np.array(y_pil)
        y_ans = [0, 0, 0, 11, 71, 200, 5, 0, 214, 31, 255, 0]
        y_ans = np.array(y_ans, dtype=np.uint8).reshape(x_shape)
        assert np.allclose(y_np, y_ans)

    def test_adjusts_L_mode(self):
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_rgb = Image.fromarray(x_np, mode='RGB')

        x_l = x_rgb.convert('L')
655
656
657
658
659
        assert F.adjust_brightness(x_l, 2).mode == 'L'
        assert F.adjust_saturation(x_l, 2).mode == 'L'
        assert F.adjust_contrast(x_l, 2).mode == 'L'
        assert F.adjust_hue(x_l, 0.4).mode == 'L'
        assert F.adjust_gamma(x_l, 0.5).mode == 'L'
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676

    def test_color_jitter(self):
        color_jitter = transforms.ColorJitter(2, 2, 2, 0.1)

        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')
        x_pil_2 = x_pil.convert('L')

        for i in range(10):
            y_pil = color_jitter(x_pil)
            assert y_pil.mode == x_pil.mode

            y_pil_2 = color_jitter(x_pil_2)
            assert y_pil_2.mode == x_pil_2.mode

677
678
679
        # Checking if ColorJitter can be printed as string
        color_jitter.__repr__()

680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
    def test_linear_transformation(self):
        x = torch.randn(250, 10, 10, 3)
        flat_x = x.view(x.size(0), x.size(1) * x.size(2) * x.size(3))
        # compute principal components
        sigma = torch.mm(flat_x.t(), flat_x) / flat_x.size(0)
        u, s, _ = np.linalg.svd(sigma.numpy())
        zca_epsilon = 1e-10  # avoid division by 0
        d = torch.Tensor(np.diag(1. / np.sqrt(s + zca_epsilon)))
        u = torch.Tensor(u)
        principal_components = torch.mm(torch.mm(u, d), u.t())
        # initialize whitening matrix
        whitening = transforms.LinearTransformation(principal_components)
        # pass first vector
        xwhite = whitening(x[0].view(10, 10, 3))
        # estimate covariance
        xwhite = xwhite.view(1, 300).numpy()
        cov = np.dot(xwhite, xwhite.T) / x.size(0)
        assert np.allclose(cov, np.identity(1), rtol=1e-3)

699
700
701
        # Checking if LinearTransformation can be printed as string
        whitening.__repr__()

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
    def test_rotate(self):
        x = np.zeros((100, 100, 3), dtype=np.uint8)
        x[40, 40] = [255, 255, 255]

        with self.assertRaises(TypeError):
            F.rotate(x, 10)

        img = F.to_pil_image(x)

        result = F.rotate(img, 45)
        assert result.size == (100, 100)
        r, c, ch = np.where(result)
        assert all(x in r for x in [49, 50])
        assert all(x in c for x in [36])
        assert all(x in ch for x in [0, 1, 2])

        result = F.rotate(img, 45, expand=True)
        assert result.size == (142, 142)
        r, c, ch = np.where(result)
        assert all(x in r for x in [70, 71])
        assert all(x in c for x in [57])
        assert all(x in ch for x in [0, 1, 2])

        result = F.rotate(img, 45, center=(40, 40))
        assert result.size == (100, 100)
        r, c, ch = np.where(result)
        assert all(x in r for x in [40])
        assert all(x in c for x in [40])
        assert all(x in ch for x in [0, 1, 2])

        result_a = F.rotate(img, 90)
        result_b = F.rotate(img, -270)

        assert np.all(np.array(result_a) == np.array(result_b))

    def test_random_rotation(self):

        with self.assertRaises(ValueError):
            transforms.RandomRotation(-0.7)
            transforms.RandomRotation([-0.7])
            transforms.RandomRotation([-0.7, 0, 0.7])

        t = transforms.RandomRotation(10)
        angle = t.get_params(t.degrees)
        assert angle > -10 and angle < 10

        t = transforms.RandomRotation((-10, 10))
        angle = t.get_params(t.degrees)
        assert angle > -10 and angle < 10

752
753
754
        # Checking if RandomRotation can be printed as string
        t.__repr__()

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
    def test_to_grayscale(self):
        """Unit tests for grayscale transform"""

        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')
        x_pil_2 = x_pil.convert('L')
        gray_np = np.array(x_pil_2)

        # Test Set: Grayscale an image with desired number of output channels
        # Case 1: RGB -> 1 channel grayscale
        trans1 = transforms.Grayscale(num_output_channels=1)
        gray_pil_1 = trans1(x_pil)
        gray_np_1 = np.array(gray_pil_1)
        assert gray_pil_1.mode == 'L', 'mode should be L'
        assert gray_np_1.shape == tuple(x_shape[0:2]), 'should be 1 channel'
        np.testing.assert_equal(gray_np, gray_np_1)

        # Case 2: RGB -> 3 channel grayscale
        trans2 = transforms.Grayscale(num_output_channels=3)
        gray_pil_2 = trans2(x_pil)
        gray_np_2 = np.array(gray_pil_2)
        assert gray_pil_2.mode == 'RGB', 'mode should be RGB'
        assert gray_np_2.shape == tuple(x_shape), 'should be 3 channel'
        np.testing.assert_equal(gray_np_2[:, :, 0], gray_np_2[:, :, 1])
        np.testing.assert_equal(gray_np_2[:, :, 1], gray_np_2[:, :, 2])
        np.testing.assert_equal(gray_np, gray_np_2[:, :, 0])

        # Case 3: 1 channel grayscale -> 1 channel grayscale
        trans3 = transforms.Grayscale(num_output_channels=1)
        gray_pil_3 = trans3(x_pil_2)
        gray_np_3 = np.array(gray_pil_3)
        assert gray_pil_3.mode == 'L', 'mode should be L'
        assert gray_np_3.shape == tuple(x_shape[0:2]), 'should be 1 channel'
        np.testing.assert_equal(gray_np, gray_np_3)

        # Case 4: 1 channel grayscale -> 3 channel grayscale
        trans4 = transforms.Grayscale(num_output_channels=3)
        gray_pil_4 = trans4(x_pil_2)
        gray_np_4 = np.array(gray_pil_4)
        assert gray_pil_4.mode == 'RGB', 'mode should be RGB'
        assert gray_np_4.shape == tuple(x_shape), 'should be 3 channel'
        np.testing.assert_equal(gray_np_4[:, :, 0], gray_np_4[:, :, 1])
        np.testing.assert_equal(gray_np_4[:, :, 1], gray_np_4[:, :, 2])
        np.testing.assert_equal(gray_np, gray_np_4[:, :, 0])

802
803
804
        # Checking if Grayscale can be printed as string
        trans4.__repr__()

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
    @unittest.skipIf(stats is None, 'scipy.stats not available')
    def test_random_grayscale(self):
        """Unit tests for random grayscale transform"""

        # Test Set 1: RGB -> 3 channel grayscale
        random_state = random.getstate()
        random.seed(42)
        x_shape = [2, 2, 3]
        x_np = np.random.randint(0, 256, x_shape, np.uint8)
        x_pil = Image.fromarray(x_np, mode='RGB')
        x_pil_2 = x_pil.convert('L')
        gray_np = np.array(x_pil_2)

        num_samples = 250
        num_gray = 0
        for _ in range(num_samples):
            gray_pil_2 = transforms.RandomGrayscale(p=0.5)(x_pil)
            gray_np_2 = np.array(gray_pil_2)
            if np.array_equal(gray_np_2[:, :, 0], gray_np_2[:, :, 1]) and \
               np.array_equal(gray_np_2[:, :, 1], gray_np_2[:, :, 2]) and \
               np.array_equal(gray_np, gray_np_2[:, :, 0]):
                num_gray = num_gray + 1

        p_value = stats.binom_test(num_gray, num_samples, p=0.5)
        random.setstate(random_state)
        assert p_value > 0.0001

        # Test Set 2: grayscale -> 1 channel grayscale
        random_state = random.getstate()
        random.seed(42)
        x_shape = [2, 2, 3]
        x_np = np.random.randint(0, 256, x_shape, np.uint8)
        x_pil = Image.fromarray(x_np, mode='RGB')
        x_pil_2 = x_pil.convert('L')
        gray_np = np.array(x_pil_2)

        num_samples = 250
        num_gray = 0
        for _ in range(num_samples):
            gray_pil_3 = transforms.RandomGrayscale(p=0.5)(x_pil_2)
            gray_np_3 = np.array(gray_pil_3)
            if np.array_equal(gray_np, gray_np_3):
                num_gray = num_gray + 1

        p_value = stats.binom_test(num_gray, num_samples, p=1.0)  # Note: grayscale is always unchanged
        random.setstate(random_state)
        assert p_value > 0.0001

        # Test set 3: Explicit tests
        x_shape = [2, 2, 3]
        x_data = [0, 5, 13, 54, 135, 226, 37, 8, 234, 90, 255, 1]
        x_np = np.array(x_data, dtype=np.uint8).reshape(x_shape)
        x_pil = Image.fromarray(x_np, mode='RGB')
        x_pil_2 = x_pil.convert('L')
        gray_np = np.array(x_pil_2)

        # Case 3a: RGB -> 3 channel grayscale (grayscaled)
        trans2 = transforms.RandomGrayscale(p=1.0)
        gray_pil_2 = trans2(x_pil)
        gray_np_2 = np.array(gray_pil_2)
        assert gray_pil_2.mode == 'RGB', 'mode should be RGB'
        assert gray_np_2.shape == tuple(x_shape), 'should be 3 channel'
        np.testing.assert_equal(gray_np_2[:, :, 0], gray_np_2[:, :, 1])
        np.testing.assert_equal(gray_np_2[:, :, 1], gray_np_2[:, :, 2])
        np.testing.assert_equal(gray_np, gray_np_2[:, :, 0])

        # Case 3b: RGB -> 3 channel grayscale (unchanged)
        trans2 = transforms.RandomGrayscale(p=0.0)
        gray_pil_2 = trans2(x_pil)
        gray_np_2 = np.array(gray_pil_2)
        assert gray_pil_2.mode == 'RGB', 'mode should be RGB'
        assert gray_np_2.shape == tuple(x_shape), 'should be 3 channel'
        np.testing.assert_equal(x_np, gray_np_2)

        # Case 3c: 1 channel grayscale -> 1 channel grayscale (grayscaled)
        trans3 = transforms.RandomGrayscale(p=1.0)
        gray_pil_3 = trans3(x_pil_2)
        gray_np_3 = np.array(gray_pil_3)
        assert gray_pil_3.mode == 'L', 'mode should be L'
        assert gray_np_3.shape == tuple(x_shape[0:2]), 'should be 1 channel'
        np.testing.assert_equal(gray_np, gray_np_3)

        # Case 3d: 1 channel grayscale -> 1 channel grayscale (unchanged)
        trans3 = transforms.RandomGrayscale(p=0.0)
        gray_pil_3 = trans3(x_pil_2)
        gray_np_3 = np.array(gray_pil_3)
        assert gray_pil_3.mode == 'L', 'mode should be L'
        assert gray_np_3.shape == tuple(x_shape[0:2]), 'should be 1 channel'
        np.testing.assert_equal(gray_np, gray_np_3)

895
896
897
        # Checking if RandomGrayscale can be printed as string
        trans3.__repr__()

898

899
900
if __name__ == '__main__':
    unittest.main()