def test_kinetics():
kinetics = Kinetics('/data/public/rw/datasets/videos/kinetics')
assert kinetics.size() == 391782
assert kinetics.names[:5] == [
'---0dWlqevI', '---QUuC4vJs', '---aQ-tA5_A', '---j12rm3WI',
'--0NTAs-fA0'
]
assert kinetics.names[-5:] == [
'zzyHRXPQP2I', 'zzyxMjfYpL0', 'zzz0-zDYts8', 'zzzZycxdZHk',
'zzz_3yWpTXo'
]
def dataflow(name='davis', scale=1, split='val'):
if name == 'davis':
ds = Davis('/data/zubin/videos/davis', name=split, num_frames=1, shuffle=False)
elif name == 'kinetics':
ds = Kinetics('/data/public/rw/datasets/videos/kinetics', num_frames=1, skips=[0], shuffle=False)
else:
raise Exception('not support dataset %s' % name)
if name != 'davis':
ds = df.MapData(ds, lambda dp: [dp[0], dp[1], dp[1]])
ds = df.MapData(ds, lambda dp: [
dp[0], # index
dp[1], # original
dp[2], # mask
dp[3], # name
])
feature_size = int(256 * scale)
size = (feature_size, feature_size)
ds = df.MapDataComponent(ds, ImageProcess.resize(small_axis=feature_size), index=1)
ds = df.MapDataComponent(ds, lambda images: cv2.resize(images[0], size), index=2)
ds = df.MapData(ds, lambda dp: [
dp[0], # index
dp[1][0], # original small axis 256 x scale
cv2.cvtColor(cv2.resize(dp[1][0], size), cv2.COLOR_BGR2GRAY).reshape((size[0], size[1], 1)), # gray (256xscale)x(256xscale)x1
dp[2], # annotated mask 256xscale x 256xscale
dp[3], # name
])
ds = df.MultiProcessPrefetchData(ds, nr_prefetch=32, nr_proc=1)
return ds
def test_kinetics_tensorpack_dataflow():
ds = Kinetics('/data/public/rw/datasets/videos/kinetics',
num_frames=4,
skips=[0, 4, 4, 8])
ds = df.MapDataComponent(
ds,
lambda images: [cv2.resize(image, (256, 256)) for image in images],
index=1)
ds = df.MapDataComponent(ds,
lambda images: np.stack(images, axis=0),
index=1)
ds = df.BatchData(ds, 6)
ds.reset_state()
generator = ds.get_data()
for _ in range(10):
_, images = next(generator)
assert images.shape == (6, 4, 256, 256, 3)
def test_kinetics_generator_with_num_frames_skips():
kinetics = Kinetics('/data/public/rw/datasets/videos/kinetics')
generator = kinetics.get_data(num_frames=4, skips=[0, 4, 4, 8])
idx, images = next(generator)
assert idx == 0
assert len(images) == 4
assert [image.shape for image in images] == [(720, 406, 3)] * 4
for _ in range(299 // (1 + 5 + 5 + 9) - 1):
next(generator)
idx, images = next(generator)
assert idx == 14
assert len(images) == 4
assert [image.shape for image in images] == [(720, 406, 3)] * 4
idx, images = next(generator)
assert idx == 0
assert len(images) == 4
assert [image.shape for image in images] == [(240, 320, 3)] * 4
def test_kinetics_generator():
kinetics = Kinetics('/data/public/rw/datasets/videos/kinetics')
generator = kinetics.get_data()
idx, images = next(generator)
assert idx == 0
assert len(images) == 1
assert images[0].shape == (720, 406, 3)
for _ in range(299 - 1):
next(generator)
idx, images = next(generator)
assert idx == 299
assert len(images) == 1
assert images[0].shape == (720, 406, 3)
idx, images = next(generator)
assert idx == 0
assert len(images) == 1
assert images[0].shape == (240, 320, 3)
def test_kinetics_get():
kinetics = Kinetics('/data/public/rw/datasets/videos/kinetics')
assert kinetics.get_filename(kinetics.names[0]) == (
True,
'/data/public/rw/datasets/videos/kinetics/processed/---0dWlqevI.mp4')
assert kinetics.get_filename(kinetics.names[-1]) == (
False,
'/data/public/rw/datasets/videos/kinetics/processed/zzz_3yWpTXo.mp4')
with pytest.raises(KeyError) as exception_info:
kinetics.get_filename('NOT_EXISTS_NAME')
assert exception_info.value.args[0] == 'not exists name at NOT_EXISTS_NAME'
def dataflow(name='davis', scale=1):
if name == 'davis':
ds = Davis('/data/public/rw/datasets/videos/davis/trainval',
num_frames=1,
shuffle=False)
elif name == 'kinetics':
ds = Kinetics('/data/public/rw/datasets/videos/kinetics',
num_frames=1,
skips=[0],
shuffle=False)
else:
raise Exception('not support dataset %s' % name)
if name != 'davis':
ds = df.MapData(ds, lambda dp: [dp[0], dp[1], dp[1]])
ds = df.MapData(
ds,
lambda dp: [
dp[0], # index
dp[1], # original
dp[2], # mask
])
size = (256 * scale, 256 * scale)
ds = df.MapDataComponent(ds,
ImageProcess.resize(small_axis=256 * scale),
index=1)
ds = df.MapDataComponent(ds,
lambda images: cv2.resize(images[0], size),
index=2)
ds = df.MapData(
ds,
lambda dp: [
dp[0], # index
dp[1][0], # original
cv2.cvtColor(cv2.resize(dp[1][0], size), cv2.COLOR_BGR2GRAY).
reshape((size[0], size[1], 1)), # gray
dp[2], # mask
])
ds = df.MultiProcessPrefetchData(ds, nr_prefetch=32, nr_proc=1)
return ds
parser.add_argument('--name', type=str, default='kinetics')
parser.add_argument('--log-filename', type=str, default='')
parser.add_argument('--debug', action='store_true')
args = parser.parse_args()
log_format = '[%(asctime)s %(levelname)s] %(message)s'
level = logging.DEBUG if args.debug else logging.INFO
if not args.log_filename:
logging.basicConfig(level=level, format=log_format, stream=sys.stderr)
else:
logging.basicConfig(level=level, format=log_format, filename=args.log_filename)
logging.info('args: %s', args)
if args.name == 'kinetics':
ds = Kinetics('/data/public/rw/datasets/videos/kinetics', num_frames=1, skips=[0], shuffle=True)
ds = df.MapData(ds, lambda dp: [dp[1][0], dp[0]])
else:
ds = df.dataset.Cifar10('train', shuffle=True)
ds = df.MapDataComponent(ds, lambda image: cv2.resize(image, (32, 32)))
ds = df.MapDataComponent(ds, lambda image: cv2.cvtColor(np.float32(image / 255.0), cv2.COLOR_RGB2Lab))
ds = df.MapDataComponent(ds, lambda image: image[:, :, 1:])
ds = df.MapDataComponent(ds, lambda image: image.reshape((-1, 2)))
ds = df.RepeatedData(ds, -1)
ds.reset_state()
generator = ds.get_data()
logging.info('initalized preprocessor')
samples = []