test_functional_tensor.py 50 KB
Newer Older
1
import colorsys
2
import itertools
3
import math
4
import os
5
import re
6
from functools import partial
7
from typing import Sequence
8

vfdev's avatar
vfdev committed
9
import numpy as np
10
import pytest
vfdev's avatar
vfdev committed
11
import torch
12
import torchvision.transforms as T
13
14
15
import torchvision.transforms.functional as F
import torchvision.transforms.functional_pil as F_pil
import torchvision.transforms.functional_tensor as F_t
Nicolas Hug's avatar
Nicolas Hug committed
16
17
18
19
20
21
22
23
from common_utils import (
    cpu_and_gpu,
    needs_cuda,
    _create_data,
    _create_data_batch,
    _assert_equal_tensor_to_pil,
    _assert_approx_equal_tensor_to_pil,
    _test_fn_on_batch,
24
    assert_equal,
Nicolas Hug's avatar
Nicolas Hug committed
25
)
26
from torchvision.transforms import InterpolationMode
27

28
NEAREST, BILINEAR, BICUBIC = InterpolationMode.NEAREST, InterpolationMode.BILINEAR, InterpolationMode.BICUBIC
29
30


31
32
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("fn", [F.get_image_size, F.get_image_num_channels])
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
def test_image_sizes(device, fn):
    script_F = torch.jit.script(fn)

    img_tensor, pil_img = _create_data(16, 18, 3, device=device)
    value_img = fn(img_tensor)
    value_pil_img = fn(pil_img)
    assert value_img == value_pil_img

    value_img_script = script_F(img_tensor)
    assert value_img == value_img_script

    batch_tensors = _create_data_batch(16, 18, 3, num_samples=4, device=device)
    value_img_batch = fn(batch_tensors)
    assert value_img == value_img_batch


49
50
51
52
53
54
55
56
@needs_cuda
def test_scale_channel():
    """Make sure that _scale_channel gives the same results on CPU and GPU as
    histc or bincount are used depending on the device.
    """
    # TODO: when # https://github.com/pytorch/pytorch/issues/53194 is fixed,
    # only use bincount and remove that test.
    size = (1_000,)
57
    img_chan = torch.randint(0, 256, size=size).to("cpu")
58
    scaled_cpu = F_t._scale_channel(img_chan)
59
60
    scaled_cuda = F_t._scale_channel(img_chan.to("cuda"))
    assert_equal(scaled_cpu, scaled_cuda.to("cpu"))
61

62

63
64
65
66
67
68
class TestRotate:

    ALL_DTYPES = [None, torch.float32, torch.float64, torch.float16]
    scripted_rotate = torch.jit.script(F.rotate)
    IMG_W = 26

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
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(26, IMG_W), (32, IMG_W)])
    @pytest.mark.parametrize(
        "center",
        [
            None,
            (int(IMG_W * 0.3), int(IMG_W * 0.4)),
            [int(IMG_W * 0.5), int(IMG_W * 0.6)],
        ],
    )
    @pytest.mark.parametrize("dt", ALL_DTYPES)
    @pytest.mark.parametrize("angle", range(-180, 180, 17))
    @pytest.mark.parametrize("expand", [True, False])
    @pytest.mark.parametrize(
        "fill",
        [
            None,
            [0, 0, 0],
            (1, 2, 3),
            [255, 255, 255],
            [
                1,
            ],
            (2.0,),
        ],
    )
    @pytest.mark.parametrize("fn", [F.rotate, scripted_rotate])
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
    def test_rotate(self, device, height, width, center, dt, angle, expand, fill, fn):
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and torch.device(device).type == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        f_pil = int(fill[0]) if fill is not None and len(fill) == 1 else fill
        out_pil_img = F.rotate(pil_img, angle=angle, interpolation=NEAREST, expand=expand, center=center, fill=f_pil)
        out_pil_tensor = torch.from_numpy(np.array(out_pil_img).transpose((2, 0, 1)))

        out_tensor = fn(tensor, angle=angle, interpolation=NEAREST, expand=expand, center=center, fill=fill).cpu()

        if out_tensor.dtype != torch.uint8:
            out_tensor = out_tensor.to(torch.uint8)

115
116
117
        assert (
            out_tensor.shape == out_pil_tensor.shape
        ), f"{(height, width, NEAREST, dt, angle, expand, center)}: {out_tensor.shape} vs {out_pil_tensor.shape}"
118
119
120
121
122
123
124

        num_diff_pixels = (out_tensor != out_pil_tensor).sum().item() / 3.0
        ratio_diff_pixels = num_diff_pixels / out_tensor.shape[-1] / out_tensor.shape[-2]
        # Tolerance : less than 3% of different pixels
        assert ratio_diff_pixels < 0.03, (
            f"{(height, width, NEAREST, dt, angle, expand, center, fill)}: "
            f"{ratio_diff_pixels}\n{out_tensor[0, :7, :7]} vs \n"
125
126
            f"{out_pil_tensor[0, :7, :7]}"
        )
127

128
129
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("dt", ALL_DTYPES)
130
131
132
133
134
135
136
137
138
139
    def test_rotate_batch(self, device, dt):
        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        batch_tensors = _create_data_batch(26, 36, num_samples=4, device=device)
        if dt is not None:
            batch_tensors = batch_tensors.to(dtype=dt)

        center = (20, 22)
140
        _test_fn_on_batch(batch_tensors, F.rotate, angle=32, interpolation=NEAREST, expand=True, center=center)
141
142
143
144

    def test_rotate_deprecation_resample(self):
        tensor, _ = _create_data(26, 26)
        # assert deprecation warning and non-BC
145
146
147
148
149
150
151
        with pytest.warns(
            UserWarning,
            match=re.escape(
                "The parameter 'resample' is deprecated since 0.12 and will be removed 0.14. "
                "Please use 'interpolation' instead."
            ),
        ):
152
153
154
155
156
157
158
159
160
161
162
163
164
            res1 = F.rotate(tensor, 45, resample=2)
            res2 = F.rotate(tensor, 45, interpolation=BILINEAR)
            assert_equal(res1, res2)

    def test_rotate_interpolation_type(self):
        tensor, _ = _create_data(26, 26)
        # assert changed type warning
        with pytest.warns(UserWarning, match=r"Argument interpolation should be of type InterpolationMode"):
            res1 = F.rotate(tensor, 45, interpolation=2)
            res2 = F.rotate(tensor, 45, interpolation=BILINEAR)
            assert_equal(res1, res2)


165
166
167
168
169
class TestAffine:

    ALL_DTYPES = [None, torch.float32, torch.float64, torch.float16]
    scripted_affine = torch.jit.script(F.affine)

170
171
172
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(26, 26), (32, 26)])
    @pytest.mark.parametrize("dt", ALL_DTYPES)
173
174
175
176
177
178
179
180
181
182
183
184
185
186
    def test_identity_map(self, device, height, width, dt):
        # Tests on square and rectangular images
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        # 1) identity map
        out_tensor = F.affine(tensor, angle=0, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST)

187
        assert_equal(tensor, out_tensor, msg=f"{out_tensor[0, :5, :5]} vs {tensor[0, :5, :5]}")
188
189
190
        out_tensor = self.scripted_affine(
            tensor, angle=0, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST
        )
191
        assert_equal(tensor, out_tensor, msg=f"{out_tensor[0, :5, :5]} vs {tensor[0, :5, :5]}")
192

193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(26, 26)])
    @pytest.mark.parametrize("dt", ALL_DTYPES)
    @pytest.mark.parametrize(
        "angle, config",
        [
            (90, {"k": 1, "dims": (-1, -2)}),
            (45, None),
            (30, None),
            (-30, None),
            (-45, None),
            (-90, {"k": -1, "dims": (-1, -2)}),
            (180, {"k": 2, "dims": (-1, -2)}),
        ],
    )
    @pytest.mark.parametrize("fn", [F.affine, scripted_affine])
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
    def test_square_rotations(self, device, height, width, dt, angle, config, fn):
        # 2) Test rotation
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        out_pil_img = F.affine(
            pil_img, angle=angle, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST
        )
        out_pil_tensor = torch.from_numpy(np.array(out_pil_img).transpose((2, 0, 1))).to(device)

225
        out_tensor = fn(tensor, angle=angle, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST)
226
        if config is not None:
227
            assert_equal(torch.rot90(tensor, **config), out_tensor)
228
229
230
231
232
233
234

        if out_tensor.dtype != torch.uint8:
            out_tensor = out_tensor.to(torch.uint8)

        num_diff_pixels = (out_tensor != out_pil_tensor).sum().item() / 3.0
        ratio_diff_pixels = num_diff_pixels / out_tensor.shape[-1] / out_tensor.shape[-2]
        # Tolerance : less than 6% of different pixels
235
        assert ratio_diff_pixels < 0.06
236

237
238
239
240
241
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(32, 26)])
    @pytest.mark.parametrize("dt", ALL_DTYPES)
    @pytest.mark.parametrize("angle", [90, 45, 15, -30, -60, -120])
    @pytest.mark.parametrize("fn", [F.affine, scripted_affine])
242
243
    @pytest.mark.parametrize("center", [None, [0, 0]])
    def test_rect_rotations(self, device, height, width, dt, angle, fn, center):
244
245
246
247
248
249
250
251
252
253
254
        # Tests on rectangular images
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        out_pil_img = F.affine(
255
            pil_img, angle=angle, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST, center=center
256
257
258
        )
        out_pil_tensor = torch.from_numpy(np.array(out_pil_img).transpose((2, 0, 1)))

259
260
261
        out_tensor = fn(
            tensor, angle=angle, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST, center=center
        ).cpu()
262
263
264
265
266
267
268

        if out_tensor.dtype != torch.uint8:
            out_tensor = out_tensor.to(torch.uint8)

        num_diff_pixels = (out_tensor != out_pil_tensor).sum().item() / 3.0
        ratio_diff_pixels = num_diff_pixels / out_tensor.shape[-1] / out_tensor.shape[-2]
        # Tolerance : less than 3% of different pixels
269
        assert ratio_diff_pixels < 0.03
270

271
272
273
274
275
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(26, 26), (32, 26)])
    @pytest.mark.parametrize("dt", ALL_DTYPES)
    @pytest.mark.parametrize("t", [[10, 12], (-12, -13)])
    @pytest.mark.parametrize("fn", [F.affine, scripted_affine])
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
    def test_translations(self, device, height, width, dt, t, fn):
        # 3) Test translation
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        out_pil_img = F.affine(pil_img, angle=0, translate=t, scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST)

        out_tensor = fn(tensor, angle=0, translate=t, scale=1.0, shear=[0.0, 0.0], interpolation=NEAREST)

        if out_tensor.dtype != torch.uint8:
            out_tensor = out_tensor.to(torch.uint8)

        _assert_equal_tensor_to_pil(out_tensor, out_pil_img)

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
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("height, width", [(26, 26), (32, 26)])
    @pytest.mark.parametrize("dt", ALL_DTYPES)
    @pytest.mark.parametrize(
        "a, t, s, sh, f",
        [
            (45.5, [5, 6], 1.0, [0.0, 0.0], None),
            (33, (5, -4), 1.0, [0.0, 0.0], [0, 0, 0]),
            (45, [-5, 4], 1.2, [0.0, 0.0], (1, 2, 3)),
            (33, (-4, -8), 2.0, [0.0, 0.0], [255, 255, 255]),
            (
                85,
                (10, -10),
                0.7,
                [0.0, 0.0],
                [
                    1,
                ],
            ),
            (
                0,
                [0, 0],
                1.0,
                [
                    35.0,
                ],
                (2.0,),
            ),
            (-25, [0, 0], 1.2, [0.0, 15.0], None),
            (-45, [-10, 0], 0.7, [2.0, 5.0], None),
            (-45, [-10, -10], 1.2, [4.0, 5.0], None),
            (-90, [0, 0], 1.0, [0.0, 0.0], None),
        ],
    )
    @pytest.mark.parametrize("fn", [F.affine, scripted_affine])
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
    def test_all_ops(self, device, height, width, dt, a, t, s, sh, f, fn):
        # 4) Test rotation + translation + scale + shear
        tensor, pil_img = _create_data(height, width, device=device)

        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        if dt is not None:
            tensor = tensor.to(dtype=dt)

        f_pil = int(f[0]) if f is not None and len(f) == 1 else f
        out_pil_img = F.affine(pil_img, angle=a, translate=t, scale=s, shear=sh, interpolation=NEAREST, fill=f_pil)
        out_pil_tensor = torch.from_numpy(np.array(out_pil_img).transpose((2, 0, 1)))

        out_tensor = fn(tensor, angle=a, translate=t, scale=s, shear=sh, interpolation=NEAREST, fill=f).cpu()

        if out_tensor.dtype != torch.uint8:
            out_tensor = out_tensor.to(torch.uint8)

        num_diff_pixels = (out_tensor != out_pil_tensor).sum().item() / 3.0
        ratio_diff_pixels = num_diff_pixels / out_tensor.shape[-1] / out_tensor.shape[-2]
        # Tolerance : less than 5% (cpu), 6% (cuda) of different pixels
        tol = 0.06 if device == "cuda" else 0.05
355
        assert ratio_diff_pixels < tol
356

357
358
    @pytest.mark.parametrize("device", cpu_and_gpu())
    @pytest.mark.parametrize("dt", ALL_DTYPES)
359
360
361
362
363
364
365
366
367
    def test_batches(self, device, dt):
        if dt == torch.float16 and device == "cpu":
            # skip float16 on CPU case
            return

        batch_tensors = _create_data_batch(26, 36, num_samples=4, device=device)
        if dt is not None:
            batch_tensors = batch_tensors.to(dtype=dt)

368
        _test_fn_on_batch(batch_tensors, F.affine, angle=-43, translate=[-3, 4], scale=1.2, shear=[4.0, 5.0])
369

370
    @pytest.mark.parametrize("device", cpu_and_gpu())
371
372
373
374
    def test_warnings(self, device):
        tensor, pil_img = _create_data(26, 26, device=device)

        # assert deprecation warning and non-BC
375
376
377
378
379
380
381
        with pytest.warns(
            UserWarning,
            match=re.escape(
                "The parameter 'resample' is deprecated since 0.12 and will be removed in 0.14. "
                "Please use 'interpolation' instead."
            ),
        ):
382
383
384
385
386
387
388
389
390
391
            res1 = F.affine(tensor, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], resample=2)
            res2 = F.affine(tensor, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=BILINEAR)
            assert_equal(res1, res2)

        # assert changed type warning
        with pytest.warns(UserWarning, match=r"Argument interpolation should be of type InterpolationMode"):
            res1 = F.affine(tensor, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=2)
            res2 = F.affine(tensor, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], interpolation=BILINEAR)
            assert_equal(res1, res2)

392
393
394
395
396
397
398
        with pytest.warns(
            UserWarning,
            match=re.escape(
                "The parameter 'fillcolor' is deprecated since 0.12 and will be removed in 0.14. "
                "Please use 'fill' instead."
            ),
        ):
399
400
401
402
403
404
            res1 = F.affine(pil_img, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], fillcolor=10)
            res2 = F.affine(pil_img, 45, translate=[0, 0], scale=1.0, shear=[0.0, 0.0], fill=10)
            # we convert the PIL images to numpy as assert_equal doesn't work on PIL images.
            assert_equal(np.asarray(res1), np.asarray(res2))


405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
def _get_data_dims_and_points_for_perspective():
    # Ideally we would parametrize independently over data dims and points, but
    # we want to tests on some points that also depend on the data dims.
    # Pytest doesn't support covariant parametrization, so we do it somewhat manually here.

    data_dims = [(26, 34), (26, 26)]
    points = [
        [[[0, 0], [33, 0], [33, 25], [0, 25]], [[3, 2], [32, 3], [30, 24], [2, 25]]],
        [[[3, 2], [32, 3], [30, 24], [2, 25]], [[0, 0], [33, 0], [33, 25], [0, 25]]],
        [[[3, 2], [32, 3], [30, 24], [2, 25]], [[5, 5], [30, 3], [33, 19], [4, 25]]],
    ]

    dims_and_points = list(itertools.product(data_dims, points))

    # up to here, we could just have used 2 @parametrized.
    # Down below is the covarariant part as the points depend on the data dims.

    n = 10
    for dim in data_dims:
424
        points += [(dim, T.RandomPerspective.get_params(dim[1], dim[0], i / n)) for i in range(n)]
425
426
427
    return dims_and_points


428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dims_and_points", _get_data_dims_and_points_for_perspective())
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize(
    "fill",
    (
        None,
        [0, 0, 0],
        [1, 2, 3],
        [255, 255, 255],
        [
            1,
        ],
        (2.0,),
    ),
)
@pytest.mark.parametrize("fn", [F.perspective, torch.jit.script(F.perspective)])
Nicolas Hug's avatar
Nicolas Hug committed
445
def test_perspective_pil_vs_tensor(device, dims_and_points, dt, fill, fn):
446
447
448
449
450
451
452

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    data_dims, (spoints, epoints) = dims_and_points

Nicolas Hug's avatar
Nicolas Hug committed
453
    tensor, pil_img = _create_data(*data_dims, device=device)
454
455
456
457
458
    if dt is not None:
        tensor = tensor.to(dtype=dt)

    interpolation = NEAREST
    fill_pil = int(fill[0]) if fill is not None and len(fill) == 1 else fill
459
460
461
    out_pil_img = F.perspective(
        pil_img, startpoints=spoints, endpoints=epoints, interpolation=interpolation, fill=fill_pil
    )
462
463
464
465
466
467
468
469
470
471
472
473
    out_pil_tensor = torch.from_numpy(np.array(out_pil_img).transpose((2, 0, 1)))
    out_tensor = fn(tensor, startpoints=spoints, endpoints=epoints, interpolation=interpolation, fill=fill).cpu()

    if out_tensor.dtype != torch.uint8:
        out_tensor = out_tensor.to(torch.uint8)

    num_diff_pixels = (out_tensor != out_pil_tensor).sum().item() / 3.0
    ratio_diff_pixels = num_diff_pixels / out_tensor.shape[-1] / out_tensor.shape[-2]
    # Tolerance : less than 5% of different pixels
    assert ratio_diff_pixels < 0.05


474
475
476
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dims_and_points", _get_data_dims_and_points_for_perspective())
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
Nicolas Hug's avatar
Nicolas Hug committed
477
def test_perspective_batch(device, dims_and_points, dt):
478
479
480
481
482
483
484

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    data_dims, (spoints, epoints) = dims_and_points

Nicolas Hug's avatar
Nicolas Hug committed
485
    batch_tensors = _create_data_batch(*data_dims, num_samples=4, device=device)
486
487
488
489
490
491
    if dt is not None:
        batch_tensors = batch_tensors.to(dtype=dt)

    # Ignore the equivalence between scripted and regular function on float16 cuda. The pixels at
    # the border may be entirely different due to small rounding errors.
    scripted_fn_atol = -1 if (dt == torch.float16 and device == "cuda") else 1e-8
Nicolas Hug's avatar
Nicolas Hug committed
492
    _test_fn_on_batch(
493
494
495
496
497
498
        batch_tensors,
        F.perspective,
        scripted_fn_atol=scripted_fn_atol,
        startpoints=spoints,
        endpoints=epoints,
        interpolation=NEAREST,
499
500
501
    )


Nicolas Hug's avatar
Nicolas Hug committed
502
def test_perspective_interpolation_warning():
503
504
505
506
    # assert changed type warning
    spoints = [[0, 0], [33, 0], [33, 25], [0, 25]]
    epoints = [[3, 2], [32, 3], [30, 24], [2, 25]]
    tensor = torch.randint(0, 256, (3, 26, 26))
Nicolas Hug's avatar
Nicolas Hug committed
507
    with pytest.warns(UserWarning, match="Argument interpolation should be of type InterpolationMode"):
508
509
        res1 = F.perspective(tensor, startpoints=spoints, endpoints=epoints, interpolation=2)
        res2 = F.perspective(tensor, startpoints=spoints, endpoints=epoints, interpolation=BILINEAR)
Nicolas Hug's avatar
Nicolas Hug committed
510
        assert_equal(res1, res2)
511
512


513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize(
    "size",
    [
        32,
        26,
        [
            32,
        ],
        [32, 32],
        (32, 32),
        [26, 35],
    ],
)
@pytest.mark.parametrize("max_size", [None, 34, 40, 1000])
@pytest.mark.parametrize("interpolation", [BILINEAR, BICUBIC, NEAREST])
Nicolas Hug's avatar
Nicolas Hug committed
530
def test_resize(device, dt, size, max_size, interpolation):
531
532
533
534
535
536
537
538
539
540

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    if max_size is not None and isinstance(size, Sequence) and len(size) != 1:
        return  # unsupported

    torch.manual_seed(12)
    script_fn = torch.jit.script(F.resize)
Nicolas Hug's avatar
Nicolas Hug committed
541
542
    tensor, pil_img = _create_data(26, 36, device=device)
    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
543
544
545
546
547
548
549
550
551
552
553

    if dt is not None:
        # This is a trivial cast to float of uint8 data to test all cases
        tensor = tensor.to(dt)
        batch_tensors = batch_tensors.to(dt)

    resized_tensor = F.resize(tensor, size=size, interpolation=interpolation, max_size=max_size)
    resized_pil_img = F.resize(pil_img, size=size, interpolation=interpolation, max_size=max_size)

    assert resized_tensor.size()[1:] == resized_pil_img.size[::-1]

554
555
556
    if interpolation not in [
        NEAREST,
    ]:
557
558
559
560
561
562
563
564
565
        # We can not check values if mode = NEAREST, as results are different
        # E.g. resized_tensor  = [[a, a, b, c, d, d, e, ...]]
        # E.g. resized_pil_img = [[a, b, c, c, d, e, f, ...]]
        resized_tensor_f = resized_tensor
        # we need to cast to uint8 to compare with PIL image
        if resized_tensor_f.dtype == torch.uint8:
            resized_tensor_f = resized_tensor_f.to(torch.float)

        # Pay attention to high tolerance for MAE
Nicolas Hug's avatar
Nicolas Hug committed
566
        _assert_approx_equal_tensor_to_pil(resized_tensor_f, resized_pil_img, tol=8.0)
567
568

    if isinstance(size, int):
569
570
571
        script_size = [
            size,
        ]
572
573
574
    else:
        script_size = size

575
    resize_result = script_fn(tensor, size=script_size, interpolation=interpolation, max_size=max_size)
576
577
    assert_equal(resized_tensor, resize_result)

578
    _test_fn_on_batch(batch_tensors, F.resize, size=script_size, interpolation=interpolation, max_size=max_size)
579
580


581
@pytest.mark.parametrize("device", cpu_and_gpu())
Nicolas Hug's avatar
Nicolas Hug committed
582
def test_resize_asserts(device):
583

Nicolas Hug's avatar
Nicolas Hug committed
584
    tensor, pil_img = _create_data(26, 36, device=device)
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600

    # assert changed type warning
    with pytest.warns(UserWarning, match=r"Argument interpolation should be of type InterpolationMode"):
        res1 = F.resize(tensor, size=32, interpolation=2)

    res2 = F.resize(tensor, size=32, interpolation=BILINEAR)
    assert_equal(res1, res2)

    for img in (tensor, pil_img):
        exp_msg = "max_size should only be passed if size specifies the length of the smaller edge"
        with pytest.raises(ValueError, match=exp_msg):
            F.resize(img, size=(32, 34), max_size=35)
        with pytest.raises(ValueError, match="max_size = 32 must be strictly greater"):
            F.resize(img, size=32, max_size=32)


601
602
603
604
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize("size", [[96, 72], [96, 420], [420, 72]])
@pytest.mark.parametrize("interpolation", [BILINEAR, BICUBIC])
Nicolas Hug's avatar
Nicolas Hug committed
605
def test_resize_antialias(device, dt, size, interpolation):
606
607
608
609
610

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

611
    torch.manual_seed(12)
612
    script_fn = torch.jit.script(F.resize)
Nicolas Hug's avatar
Nicolas Hug committed
613
    tensor, pil_img = _create_data(320, 290, device=device)
614
615
616
617
618
619
620
621

    if dt is not None:
        # This is a trivial cast to float of uint8 data to test all cases
        tensor = tensor.to(dt)

    resized_tensor = F.resize(tensor, size=size, interpolation=interpolation, antialias=True)
    resized_pil_img = F.resize(pil_img, size=size, interpolation=interpolation)

Nicolas Hug's avatar
Nicolas Hug committed
622
    assert resized_tensor.size()[1:] == resized_pil_img.size[::-1]
623
624
625
626
627
628

    resized_tensor_f = resized_tensor
    # we need to cast to uint8 to compare with PIL image
    if resized_tensor_f.dtype == torch.uint8:
        resized_tensor_f = resized_tensor_f.to(torch.float)

629
    _assert_approx_equal_tensor_to_pil(resized_tensor_f, resized_pil_img, tol=0.5, msg=f"{size}, {interpolation}, {dt}")
630
631
632
633
634
635
636
637
638

    accepted_tol = 1.0 + 1e-5
    if interpolation == BICUBIC:
        # this overall mean value to make the tests pass
        # High value is mostly required for test cases with
        # downsampling and upsampling where we can not exactly
        # match PIL implementation.
        accepted_tol = 15.0

Nicolas Hug's avatar
Nicolas Hug committed
639
    _assert_approx_equal_tensor_to_pil(
640
        resized_tensor_f, resized_pil_img, tol=accepted_tol, agg_method="max", msg=f"{size}, {interpolation}, {dt}"
641
642
643
    )

    if isinstance(size, int):
644
645
646
        script_size = [
            size,
        ]
647
648
649
650
    else:
        script_size = size

    resize_result = script_fn(tensor, size=script_size, interpolation=interpolation, antialias=True)
Nicolas Hug's avatar
Nicolas Hug committed
651
    assert_equal(resized_tensor, resize_result)
652
653


654
@needs_cuda
655
@pytest.mark.parametrize("interpolation", [BILINEAR, BICUBIC])
Nicolas Hug's avatar
Nicolas Hug committed
656
def test_assert_resize_antialias(interpolation):
657
658

    # Checks implementation on very large scales
659
    # and catch TORCH_CHECK inside PyTorch implementation
660
    torch.manual_seed(12)
661
    tensor, _ = _create_data(1000, 1000, device="cuda")
662

663
664
665
    # Error message is not yet updated in pytorch nightly
    # with pytest.raises(RuntimeError, match=r"Provided interpolation parameters can not be handled"):
    with pytest.raises(RuntimeError, match=r"Too much shared memory required"):
666
667
668
        F.resize(tensor, size=(5, 5), interpolation=interpolation, antialias=True)


669
670
671
672
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dt", [torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize("size", [[10, 7], [10, 42], [42, 7]])
@pytest.mark.parametrize("interpolation", [BILINEAR, BICUBIC])
673
def test_interpolate_antialias_backward(device, dt, size, interpolation):
674
675
676
677
678
679

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    torch.manual_seed(12)
680
    x = (torch.rand(1, 32, 29, 3, dtype=torch.double, device=device).permute(0, 3, 1, 2).requires_grad_(True),)
681
682
    resize = partial(F.resize, size=size, interpolation=interpolation, antialias=True)
    assert torch.autograd.gradcheck(resize, x, eps=1e-8, atol=1e-6, rtol=1e-6, fast_mode=False)
683

684
    x = (torch.rand(1, 3, 32, 29, dtype=torch.double, device=device, requires_grad=True),)
685
    assert torch.autograd.gradcheck(resize, x, eps=1e-8, atol=1e-6, rtol=1e-6, fast_mode=False)
686
687


688
689
690
def check_functional_vs_PIL_vs_scripted(
    fn, fn_pil, fn_t, config, device, dtype, channels=3, tol=2.0 + 1e-10, agg_method="max"
):
691
692
693

    script_fn = torch.jit.script(fn)
    torch.manual_seed(15)
694
695
    tensor, pil_img = _create_data(26, 34, channels=channels, device=device)
    batch_tensors = _create_data_batch(16, 18, num_samples=4, channels=channels, device=device)
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713

    if dtype is not None:
        tensor = F.convert_image_dtype(tensor, dtype)
        batch_tensors = F.convert_image_dtype(batch_tensors, dtype)

    out_fn_t = fn_t(tensor, **config)
    out_pil = fn_pil(pil_img, **config)
    out_scripted = script_fn(tensor, **config)
    assert out_fn_t.dtype == out_scripted.dtype
    assert out_fn_t.size()[1:] == out_pil.size[::-1]

    rbg_tensor = out_fn_t

    if out_fn_t.dtype != torch.uint8:
        rbg_tensor = F.convert_image_dtype(out_fn_t, torch.uint8)

    # Check that max difference does not exceed 2 in [0, 255] range
    # Exact matching is not possible due to incompatibility convert_image_dtype and PIL results
Nicolas Hug's avatar
Nicolas Hug committed
714
    _assert_approx_equal_tensor_to_pil(rbg_tensor.float(), out_pil, tol=tol, agg_method=agg_method)
715
716
717
718
719
720
721

    atol = 1e-6
    if out_fn_t.dtype == torch.uint8 and "cuda" in torch.device(device).type:
        atol = 1.0
    assert out_fn_t.allclose(out_scripted, atol=atol)

    # FIXME: fn will be scripted again in _test_fn_on_batch. We could avoid that.
Nicolas Hug's avatar
Nicolas Hug committed
722
    _test_fn_on_batch(batch_tensors, fn, scripted_fn_atol=atol, **config)
723
724


725
726
727
728
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"brightness_factor": f} for f in (0.1, 0.5, 1.0, 1.34, 2.5)])
@pytest.mark.parametrize("channels", [1, 3])
729
def test_adjust_brightness(device, dtype, config, channels):
730
731
732
733
734
735
736
    check_functional_vs_PIL_vs_scripted(
        F.adjust_brightness,
        F_pil.adjust_brightness,
        F_t.adjust_brightness,
        config,
        device,
        dtype,
737
        channels,
738
739
740
    )


741
742
743
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("channels", [1, 3])
744
def test_invert(device, dtype, channels):
745
    check_functional_vs_PIL_vs_scripted(
746
        F.invert, F_pil.invert, F_t.invert, {}, device, dtype, channels, tol=1.0, agg_method="max"
747
748
749
    )


750
751
752
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("config", [{"bits": bits} for bits in range(0, 8)])
@pytest.mark.parametrize("channels", [1, 3])
753
def test_posterize(device, config, channels):
754
755
756
757
758
759
760
    check_functional_vs_PIL_vs_scripted(
        F.posterize,
        F_pil.posterize,
        F_t.posterize,
        config,
        device,
        dtype=None,
761
        channels=channels,
762
763
764
765
766
        tol=1.0,
        agg_method="max",
    )


767
768
769
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("config", [{"threshold": threshold} for threshold in [0, 64, 128, 192, 255]])
@pytest.mark.parametrize("channels", [1, 3])
770
def test_solarize1(device, config, channels):
771
772
773
774
775
776
777
    check_functional_vs_PIL_vs_scripted(
        F.solarize,
        F_pil.solarize,
        F_t.solarize,
        config,
        device,
        dtype=None,
778
        channels=channels,
779
780
781
782
783
        tol=1.0,
        agg_method="max",
    )


784
785
786
787
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"threshold": threshold} for threshold in [0.0, 0.25, 0.5, 0.75, 1.0]])
@pytest.mark.parametrize("channels", [1, 3])
788
def test_solarize2(device, dtype, config, channels):
789
790
791
792
793
794
795
    check_functional_vs_PIL_vs_scripted(
        F.solarize,
        lambda img, threshold: F_pil.solarize(img, 255 * threshold),
        F_t.solarize,
        config,
        device,
        dtype,
796
        channels,
797
798
799
800
801
        tol=1.0,
        agg_method="max",
    )


puhuk's avatar
puhuk committed
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
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("threshold", [0.0, 0.25, 0.5, 0.75, 1.0])
def test_solarize_threshold1_bound(threshold, device):
    img = torch.rand((3, 12, 23)).to(device)
    F_t.solarize(img, threshold)


@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("threshold", [1.5])
def test_solarize_threshold1_upper_bound(threshold, device):
    img = torch.rand((3, 12, 23)).to(device)
    with pytest.raises(TypeError, match="Threshold should be less than bound of img."):
        F_t.solarize(img, threshold)


@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("threshold", [0, 64, 128, 192, 255])
def test_solarize_threshold2_bound(threshold, device):
    img = torch.randint(0, 256, (3, 12, 23)).to(device)
    F_t.solarize(img, threshold)


@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("threshold", [260])
def test_solarize_threshold2_upper_bound(threshold, device):
    img = torch.randint(0, 256, (3, 12, 23)).to(device)
    with pytest.raises(TypeError, match="Threshold should be less than bound of img."):
        F_t.solarize(img, threshold)


832
833
834
835
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"sharpness_factor": f} for f in [0.2, 0.5, 1.0, 1.5, 2.0]])
@pytest.mark.parametrize("channels", [1, 3])
836
def test_adjust_sharpness(device, dtype, config, channels):
837
838
839
840
841
842
843
    check_functional_vs_PIL_vs_scripted(
        F.adjust_sharpness,
        F_pil.adjust_sharpness,
        F_t.adjust_sharpness,
        config,
        device,
        dtype,
844
        channels,
845
846
847
    )


848
849
850
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("channels", [1, 3])
851
def test_autocontrast(device, dtype, channels):
852
    check_functional_vs_PIL_vs_scripted(
853
        F.autocontrast, F_pil.autocontrast, F_t.autocontrast, {}, device, dtype, channels, tol=1.0, agg_method="max"
854
855
856
    )


857
858
859
860
861
862
863
864
865
866
867
868
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("channels", [1, 3])
def test_autocontrast_equal_minmax(device, dtype, channels):
    a = _create_data_batch(32, 32, num_samples=1, channels=channels, device=device)
    a = a / 2.0 + 0.3
    assert (F.autocontrast(a)[0] == F.autocontrast(a[0])).all()

    a[0, 0] = 0.7
    assert (F.autocontrast(a)[0] == F.autocontrast(a[0])).all()


869
870
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("channels", [1, 3])
871
def test_equalize(device, channels):
872
    torch.use_deterministic_algorithms(False)
873
874
875
876
877
878
879
    check_functional_vs_PIL_vs_scripted(
        F.equalize,
        F_pil.equalize,
        F_t.equalize,
        {},
        device,
        dtype=None,
880
        channels=channels,
881
882
883
884
885
        tol=1.0,
        agg_method="max",
    )


886
887
888
889
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"contrast_factor": f} for f in [0.2, 0.5, 1.0, 1.5, 2.0]])
@pytest.mark.parametrize("channels", [1, 3])
890
def test_adjust_contrast(device, dtype, config, channels):
891
    check_functional_vs_PIL_vs_scripted(
892
        F.adjust_contrast, F_pil.adjust_contrast, F_t.adjust_contrast, config, device, dtype, channels
893
894
895
    )


896
897
898
899
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"saturation_factor": f} for f in [0.5, 0.75, 1.0, 1.5, 2.0]])
@pytest.mark.parametrize("channels", [1, 3])
900
def test_adjust_saturation(device, dtype, config, channels):
901
    check_functional_vs_PIL_vs_scripted(
902
        F.adjust_saturation, F_pil.adjust_saturation, F_t.adjust_saturation, config, device, dtype, channels
903
904
905
    )


906
907
908
909
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"hue_factor": f} for f in [-0.45, -0.25, 0.0, 0.25, 0.45]])
@pytest.mark.parametrize("channels", [1, 3])
910
def test_adjust_hue(device, dtype, config, channels):
911
    check_functional_vs_PIL_vs_scripted(
912
        F.adjust_hue, F_pil.adjust_hue, F_t.adjust_hue, config, device, dtype, channels, tol=16.1, agg_method="max"
913
914
915
    )


916
917
918
919
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dtype", (None, torch.float32, torch.float64))
@pytest.mark.parametrize("config", [{"gamma": g1, "gain": g2} for g1, g2 in zip([0.8, 1.0, 1.2], [0.7, 1.0, 1.3])])
@pytest.mark.parametrize("channels", [1, 3])
920
def test_adjust_gamma(device, dtype, config, channels):
921
922
923
924
925
926
927
    check_functional_vs_PIL_vs_scripted(
        F.adjust_gamma,
        F_pil.adjust_gamma,
        F_t.adjust_gamma,
        config,
        device,
        dtype,
928
        channels,
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
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize(
    "pad",
    [
        2,
        [
            3,
        ],
        [0, 3],
        (3, 3),
        [4, 2, 4, 3],
    ],
)
@pytest.mark.parametrize(
    "config",
    [
        {"padding_mode": "constant", "fill": 0},
        {"padding_mode": "constant", "fill": 10},
        {"padding_mode": "constant", "fill": 20},
        {"padding_mode": "edge"},
        {"padding_mode": "reflect"},
        {"padding_mode": "symmetric"},
    ],
)
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
def test_pad(device, dt, pad, config):
    script_fn = torch.jit.script(F.pad)
    tensor, pil_img = _create_data(7, 8, device=device)
    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    if dt is not None:
        # This is a trivial cast to float of uint8 data to test all cases
        tensor = tensor.to(dt)
        batch_tensors = batch_tensors.to(dt)

    pad_tensor = F_t.pad(tensor, pad, **config)
    pad_pil_img = F_pil.pad(pil_img, pad, **config)

    pad_tensor_8b = pad_tensor
    # we need to cast to uint8 to compare with PIL image
    if pad_tensor_8b.dtype != torch.uint8:
        pad_tensor_8b = pad_tensor_8b.to(torch.uint8)

979
    _assert_equal_tensor_to_pil(pad_tensor_8b, pad_pil_img, msg=f"{pad}, {config}")
980
981

    if isinstance(pad, int):
982
983
984
        script_pad = [
            pad,
        ]
985
986
987
    else:
        script_pad = pad
    pad_tensor_script = script_fn(tensor, script_pad, **config)
988
    assert_equal(pad_tensor, pad_tensor_script, msg=f"{pad}, {config}")
989
990
991
992

    _test_fn_on_batch(batch_tensors, F.pad, padding=script_pad, **config)


993
994
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("mode", [NEAREST, BILINEAR, BICUBIC])
995
996
997
998
999
1000
def test_resized_crop(device, mode):
    # test values of F.resized_crop in several cases:
    # 1) resize to the same size, crop to the same size => should be identity
    tensor, _ = _create_data(26, 36, device=device)

    out_tensor = F.resized_crop(tensor, top=0, left=0, height=26, width=36, size=[26, 36], interpolation=mode)
1001
    assert_equal(tensor, out_tensor, msg=f"{out_tensor[0, :5, :5]} vs {tensor[0, :5, :5]}")
1002
1003
1004
1005
1006
1007
1008
1009

    # 2) resize by half and crop a TL corner
    tensor, _ = _create_data(26, 36, device=device)
    out_tensor = F.resized_crop(tensor, top=0, left=0, height=20, width=30, size=[10, 15], interpolation=NEAREST)
    expected_out_tensor = tensor[:, :20:2, :30:2]
    assert_equal(
        expected_out_tensor,
        out_tensor,
1010
        msg=f"{expected_out_tensor[0, :10, :10]} vs {out_tensor[0, :10, :10]}",
1011
1012
1013
1014
1015
1016
1017
1018
    )

    batch_tensors = _create_data_batch(26, 36, num_samples=4, device=device)
    _test_fn_on_batch(
        batch_tensors, F.resized_crop, top=1, left=2, height=20, width=30, size=[10, 15], interpolation=NEAREST
    )


1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize(
    "func, args",
    [
        (F_t.get_image_size, ()),
        (F_t.vflip, ()),
        (F_t.hflip, ()),
        (F_t.crop, (1, 2, 4, 5)),
        (F_t.adjust_brightness, (0.0,)),
        (F_t.adjust_contrast, (1.0,)),
        (F_t.adjust_hue, (-0.5,)),
        (F_t.adjust_saturation, (2.0,)),
        (
            F_t.pad,
            (
                [
                    2,
                ],
                2,
                "constant",
            ),
        ),
        (F_t.resize, ([10, 11],)),
        (
            F_t.perspective,
            (
                [
                    0.2,
                ]
            ),
        ),
        (F_t.gaussian_blur, ((2, 2), (0.7, 0.5))),
        (F_t.invert, ()),
        (F_t.posterize, (0,)),
        (F_t.solarize, (0.3,)),
        (F_t.adjust_sharpness, (0.3,)),
        (F_t.autocontrast, ()),
        (F_t.equalize, ()),
    ],
)
1059
1060
1061
1062
1063
1064
1065
def test_assert_image_tensor(device, func, args):
    shape = (100,)
    tensor = torch.rand(*shape, dtype=torch.float, device=device)
    with pytest.raises(Exception, match=r"Tensor is not a torch image."):
        func(tensor, *args)


1066
@pytest.mark.parametrize("device", cpu_and_gpu())
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
def test_vflip(device):
    script_vflip = torch.jit.script(F.vflip)

    img_tensor, pil_img = _create_data(16, 18, device=device)
    vflipped_img = F.vflip(img_tensor)
    vflipped_pil_img = F.vflip(pil_img)
    _assert_equal_tensor_to_pil(vflipped_img, vflipped_pil_img)

    # scriptable function test
    vflipped_img_script = script_vflip(img_tensor)
    assert_equal(vflipped_img, vflipped_img_script)

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    _test_fn_on_batch(batch_tensors, F.vflip)


1083
@pytest.mark.parametrize("device", cpu_and_gpu())
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
def test_hflip(device):
    script_hflip = torch.jit.script(F.hflip)

    img_tensor, pil_img = _create_data(16, 18, device=device)
    hflipped_img = F.hflip(img_tensor)
    hflipped_pil_img = F.hflip(pil_img)
    _assert_equal_tensor_to_pil(hflipped_img, hflipped_pil_img)

    # scriptable function test
    hflipped_img_script = script_hflip(img_tensor)
    assert_equal(hflipped_img, hflipped_img_script)

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    _test_fn_on_batch(batch_tensors, F.hflip)


1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize(
    "top, left, height, width",
    [
        (1, 2, 4, 5),  # crop inside top-left corner
        (2, 12, 3, 4),  # crop inside top-right corner
        (8, 3, 5, 6),  # crop inside bottom-left corner
        (8, 11, 4, 3),  # crop inside bottom-right corner
    ],
)
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
def test_crop(device, top, left, height, width):
    script_crop = torch.jit.script(F.crop)

    img_tensor, pil_img = _create_data(16, 18, device=device)

    pil_img_cropped = F.crop(pil_img, top, left, height, width)

    img_tensor_cropped = F.crop(img_tensor, top, left, height, width)
    _assert_equal_tensor_to_pil(img_tensor_cropped, pil_img_cropped)

    img_tensor_cropped = script_crop(img_tensor, top, left, height, width)
    _assert_equal_tensor_to_pil(img_tensor_cropped, pil_img_cropped)

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    _test_fn_on_batch(batch_tensors, F.crop, top=top, left=left, height=height, width=width)


1127
1128
1129
1130
1131
1132
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("image_size", ("small", "large"))
@pytest.mark.parametrize("dt", [None, torch.float32, torch.float64, torch.float16])
@pytest.mark.parametrize("ksize", [(3, 3), [3, 5], (23, 23)])
@pytest.mark.parametrize("sigma", [[0.5, 0.5], (0.5, 0.5), (0.8, 0.8), (1.7, 1.7)])
@pytest.mark.parametrize("fn", [F.gaussian_blur, torch.jit.script(F.gaussian_blur)])
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
def test_gaussian_blur(device, image_size, dt, ksize, sigma, fn):

    # true_cv2_results = {
    #     # np_img = np.arange(3 * 10 * 12, dtype="uint8").reshape((10, 12, 3))
    #     # cv2.GaussianBlur(np_img, ksize=(3, 3), sigmaX=0.8)
    #     "3_3_0.8": ...
    #     # cv2.GaussianBlur(np_img, ksize=(3, 3), sigmaX=0.5)
    #     "3_3_0.5": ...
    #     # cv2.GaussianBlur(np_img, ksize=(3, 5), sigmaX=0.8)
    #     "3_5_0.8": ...
    #     # cv2.GaussianBlur(np_img, ksize=(3, 5), sigmaX=0.5)
    #     "3_5_0.5": ...
    #     # np_img2 = np.arange(26 * 28, dtype="uint8").reshape((26, 28))
    #     # cv2.GaussianBlur(np_img2, ksize=(23, 23), sigmaX=1.7)
    #     "23_23_1.7": ...
    # }
1149
    p = os.path.join(os.path.dirname(os.path.abspath(__file__)), "assets", "gaussian_blur_opencv_results.pt")
1150
1151
    true_cv2_results = torch.load(p)

1152
1153
1154
1155
    if image_size == "small":
        tensor = (
            torch.from_numpy(np.arange(3 * 10 * 12, dtype="uint8").reshape((10, 12, 3))).permute(2, 0, 1).to(device)
        )
1156
    else:
1157
        tensor = torch.from_numpy(np.arange(26 * 28, dtype="uint8").reshape((1, 26, 28))).to(device)
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168

    if dt == torch.float16 and device == "cpu":
        # skip float16 on CPU case
        return

    if dt is not None:
        tensor = tensor.to(dtype=dt)

    _ksize = (ksize, ksize) if isinstance(ksize, int) else ksize
    _sigma = sigma[0] if sigma is not None else None
    shape = tensor.shape
1169
    gt_key = f"{shape[-2]}_{shape[-1]}_{shape[-3]}__{_ksize[0]}_{_ksize[1]}_{_sigma}"
1170
1171
1172
    if gt_key not in true_cv2_results:
        return

1173
1174
1175
    true_out = (
        torch.tensor(true_cv2_results[gt_key]).reshape(shape[-2], shape[-1], shape[-3]).permute(2, 0, 1).to(tensor)
    )
1176
1177

    out = fn(tensor, kernel_size=ksize, sigma=sigma)
1178
    torch.testing.assert_close(out, true_out, rtol=0.0, atol=1.0, msg=f"{ksize}, {sigma}")
1179
1180


1181
@pytest.mark.parametrize("device", cpu_and_gpu())
1182
1183
1184
1185
1186
1187
1188
1189
def test_hsv2rgb(device):
    scripted_fn = torch.jit.script(F_t._hsv2rgb)
    shape = (3, 100, 150)
    for _ in range(10):
        hsv_img = torch.rand(*shape, dtype=torch.float, device=device)
        rgb_img = F_t._hsv2rgb(hsv_img)
        ft_img = rgb_img.permute(1, 2, 0).flatten(0, 1)

1190
1191
1192
1193
1194
        (
            h,
            s,
            v,
        ) = hsv_img.unbind(0)
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
        h = h.flatten().cpu().numpy()
        s = s.flatten().cpu().numpy()
        v = v.flatten().cpu().numpy()

        rgb = []
        for h1, s1, v1 in zip(h, s, v):
            rgb.append(colorsys.hsv_to_rgb(h1, s1, v1))
        colorsys_img = torch.tensor(rgb, dtype=torch.float32, device=device)
        torch.testing.assert_close(ft_img, colorsys_img, rtol=0.0, atol=1e-5)

        s_rgb_img = scripted_fn(hsv_img)
        torch.testing.assert_close(rgb_img, s_rgb_img)

    batch_tensors = _create_data_batch(120, 100, num_samples=4, device=device).float()
    _test_fn_on_batch(batch_tensors, F_t._hsv2rgb)


1212
@pytest.mark.parametrize("device", cpu_and_gpu())
1213
1214
1215
1216
1217
1218
1219
1220
def test_rgb2hsv(device):
    scripted_fn = torch.jit.script(F_t._rgb2hsv)
    shape = (3, 150, 100)
    for _ in range(10):
        rgb_img = torch.rand(*shape, dtype=torch.float, device=device)
        hsv_img = F_t._rgb2hsv(rgb_img)
        ft_hsv_img = hsv_img.permute(1, 2, 0).flatten(0, 1)

1221
1222
1223
1224
1225
        (
            r,
            g,
            b,
        ) = rgb_img.unbind(dim=-3)
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
        r = r.flatten().cpu().numpy()
        g = g.flatten().cpu().numpy()
        b = b.flatten().cpu().numpy()

        hsv = []
        for r1, g1, b1 in zip(r, g, b):
            hsv.append(colorsys.rgb_to_hsv(r1, g1, b1))

        colorsys_img = torch.tensor(hsv, dtype=torch.float32, device=device)

        ft_hsv_img_h, ft_hsv_img_sv = torch.split(ft_hsv_img, [1, 2], dim=1)
        colorsys_img_h, colorsys_img_sv = torch.split(colorsys_img, [1, 2], dim=1)

        max_diff_h = ((colorsys_img_h * 2 * math.pi).sin() - (ft_hsv_img_h * 2 * math.pi).sin()).abs().max()
        max_diff_sv = (colorsys_img_sv - ft_hsv_img_sv).abs().max()
        max_diff = max(max_diff_h, max_diff_sv)
        assert max_diff < 1e-5

        s_hsv_img = scripted_fn(rgb_img)
        torch.testing.assert_close(hsv_img, s_hsv_img, rtol=1e-5, atol=1e-7)

    batch_tensors = _create_data_batch(120, 100, num_samples=4, device=device).float()
    _test_fn_on_batch(batch_tensors, F_t._rgb2hsv)


1251
1252
@pytest.mark.parametrize("device", cpu_and_gpu())
@pytest.mark.parametrize("num_output_channels", (3, 1))
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
def test_rgb_to_grayscale(device, num_output_channels):
    script_rgb_to_grayscale = torch.jit.script(F.rgb_to_grayscale)

    img_tensor, pil_img = _create_data(32, 34, device=device)

    gray_pil_image = F.rgb_to_grayscale(pil_img, num_output_channels=num_output_channels)
    gray_tensor = F.rgb_to_grayscale(img_tensor, num_output_channels=num_output_channels)

    _assert_approx_equal_tensor_to_pil(gray_tensor.float(), gray_pil_image, tol=1.0 + 1e-10, agg_method="max")

    s_gray_tensor = script_rgb_to_grayscale(img_tensor, num_output_channels=num_output_channels)
    assert_equal(s_gray_tensor, gray_tensor)

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    _test_fn_on_batch(batch_tensors, F.rgb_to_grayscale, num_output_channels=num_output_channels)


1270
@pytest.mark.parametrize("device", cpu_and_gpu())
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
def test_center_crop(device):
    script_center_crop = torch.jit.script(F.center_crop)

    img_tensor, pil_img = _create_data(32, 34, device=device)

    cropped_pil_image = F.center_crop(pil_img, [10, 11])

    cropped_tensor = F.center_crop(img_tensor, [10, 11])
    _assert_equal_tensor_to_pil(cropped_tensor, cropped_pil_image)

    cropped_tensor = script_center_crop(img_tensor, [10, 11])
    _assert_equal_tensor_to_pil(cropped_tensor, cropped_pil_image)

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    _test_fn_on_batch(batch_tensors, F.center_crop, output_size=[10, 11])


1288
@pytest.mark.parametrize("device", cpu_and_gpu())
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
def test_five_crop(device):
    script_five_crop = torch.jit.script(F.five_crop)

    img_tensor, pil_img = _create_data(32, 34, device=device)

    cropped_pil_images = F.five_crop(pil_img, [10, 11])

    cropped_tensors = F.five_crop(img_tensor, [10, 11])
    for i in range(5):
        _assert_equal_tensor_to_pil(cropped_tensors[i], cropped_pil_images[i])

    cropped_tensors = script_five_crop(img_tensor, [10, 11])
    for i in range(5):
        _assert_equal_tensor_to_pil(cropped_tensors[i], cropped_pil_images[i])

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    tuple_transformed_batches = F.five_crop(batch_tensors, [10, 11])
    for i in range(len(batch_tensors)):
        img_tensor = batch_tensors[i, ...]
        tuple_transformed_imgs = F.five_crop(img_tensor, [10, 11])
        assert len(tuple_transformed_imgs) == len(tuple_transformed_batches)

        for j in range(len(tuple_transformed_imgs)):
            true_transformed_img = tuple_transformed_imgs[j]
            transformed_img = tuple_transformed_batches[j][i, ...]
            assert_equal(true_transformed_img, transformed_img)

    # scriptable function test
    s_tuple_transformed_batches = script_five_crop(batch_tensors, [10, 11])
    for transformed_batch, s_transformed_batch in zip(tuple_transformed_batches, s_tuple_transformed_batches):
        assert_equal(transformed_batch, s_transformed_batch)


1322
@pytest.mark.parametrize("device", cpu_and_gpu())
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
def test_ten_crop(device):
    script_ten_crop = torch.jit.script(F.ten_crop)

    img_tensor, pil_img = _create_data(32, 34, device=device)

    cropped_pil_images = F.ten_crop(pil_img, [10, 11])

    cropped_tensors = F.ten_crop(img_tensor, [10, 11])
    for i in range(10):
        _assert_equal_tensor_to_pil(cropped_tensors[i], cropped_pil_images[i])

    cropped_tensors = script_ten_crop(img_tensor, [10, 11])
    for i in range(10):
        _assert_equal_tensor_to_pil(cropped_tensors[i], cropped_pil_images[i])

    batch_tensors = _create_data_batch(16, 18, num_samples=4, device=device)
    tuple_transformed_batches = F.ten_crop(batch_tensors, [10, 11])
    for i in range(len(batch_tensors)):
        img_tensor = batch_tensors[i, ...]
        tuple_transformed_imgs = F.ten_crop(img_tensor, [10, 11])
        assert len(tuple_transformed_imgs) == len(tuple_transformed_batches)

        for j in range(len(tuple_transformed_imgs)):
            true_transformed_img = tuple_transformed_imgs[j]
            transformed_img = tuple_transformed_batches[j][i, ...]
            assert_equal(true_transformed_img, transformed_img)

    # scriptable function test
    s_tuple_transformed_batches = script_ten_crop(batch_tensors, [10, 11])
    for transformed_batch, s_transformed_batch in zip(tuple_transformed_batches, s_tuple_transformed_batches):
        assert_equal(transformed_batch, s_transformed_batch)


1356
if __name__ == "__main__":
1357
    pytest.main([__file__])