def test_albu_transform():
results = dict(img_prefix=osp.join(osp.dirname(__file__), '../data'),
img_info=dict(filename='color.jpg'))
# Define simple pipeline
load = dict(type='LoadImageFromFile')
load = build_from_cfg(load, PIPELINES)
albu_transform = dict(type='Albu',
transforms=[dict(type='ChannelShuffle', p=1)])
albu_transform = build_from_cfg(albu_transform, PIPELINES)
normalize = dict(type='Normalize', mean=[0] * 3, std=[0] * 3, to_rgb=True)
normalize = build_from_cfg(normalize, PIPELINES)
# Execute transforms
results = load(results)
results = albu_transform(results)
results = normalize(results)
assert results['img'].dtype == np.float32
def __init__(self, transforms):
assert isinstance(transforms, Sequence)
self.transforms = []
for transform in transforms:
if isinstance(transform, dict):
transform = build_from_cfg(transform, PIPELINES)
self.transforms.append(transform)
elif callable(transform):
self.transforms.append(transform)
else:
raise TypeError('transform must be callable or a dict, but got'
f' {type(transform)}')
def build_activation_layer(cfg):
"""Build activation layer.
Args:
cfg (dict): The activation layer config, which should contain:
- type (str): Layer type.
- layer args: Args needed to instantiate an activation layer.
Returns:
nn.Module: Created activation layer.
"""
return build_from_cfg(cfg, ACTIVATION_LAYERS)
def build(cfg, registry, default_args=None):
"""Build a module.
Args:
cfg (dict, list[dict]): The config of modules, it is either a dict
or a list of configs.
registry (:obj:`Registry`): A registry the module belongs to.
default_args (dict, optional): Default arguments to build the module.
Defaults to None.
Returns:
nn.Module: A built nn module.
"""
if isinstance(cfg, list):
modules = [
build_from_cfg(cfg_, registry, default_args) for cfg_ in cfg
]
return nn.Sequential(*modules)
return build_from_cfg(cfg, registry, default_args)
def test_aug_test_size():
results = dict(img_prefix=osp.join(osp.dirname(__file__), '../../../data'),
img_info=dict(filename='color.jpg'))
# Define simple pipeline
load = dict(type='LoadImageFromFile')
load = build_from_cfg(load, PIPELINES)
# get config
transform = dict(type='MultiScaleFlipAug',
transforms=[],
img_scale=[(1333, 800), (800, 600), (640, 480)],
flip=True,
flip_direction=['horizontal', 'vertical', 'diagonal'])
multi_aug_test_module = build_from_cfg(transform, PIPELINES)
results = load(results)
results = multi_aug_test_module(load(results))
# len(["original", "horizontal", "vertical", "diagonal"]) *
# len([(1333, 800), (800, 600), (640, 480)])
assert len(results['img']) == 12
def test_flip():
# test assertion for invalid flip_ratio
with pytest.raises(AssertionError):
transform = dict(type='RandomFlip', flip_ratio=1.5)
build_from_cfg(transform, PIPELINES)
# test assertion for invalid direction
with pytest.raises(AssertionError):
transform = dict(
type='RandomFlip', flip_ratio=1, direction='horizonta')
build_from_cfg(transform, PIPELINES)
transform = dict(type='RandomFlip', flip_ratio=1)
flip_module = build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../data/color.jpg'), 'color')
original_img = copy.deepcopy(img)
results['img'] = img
results['img2'] = copy.deepcopy(img)
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
# Set initial values for default meta_keys
results['pad_shape'] = img.shape
results['scale_factor'] = 1.0
results['img_fields'] = ['img', 'img2']
results = flip_module(results)
assert np.equal(results['img'], results['img2']).all()
flip_module = build_from_cfg(transform, PIPELINES)
results = flip_module(results)
assert np.equal(results['img'], results['img2']).all()
assert np.equal(original_img, results['img']).all()
def build_dataset(cfg, default_args=None):
from .dataset_wrappers import ConcatDataset, RepeatDataset
if isinstance(cfg, (list, tuple)):
dataset = ConcatDataset([build_dataset(c, default_args) for c in cfg])
elif cfg['type'] == 'RepeatDataset':
dataset = RepeatDataset(build_dataset(cfg['dataset'], default_args),
cfg['times'])
elif isinstance(cfg.get('ann_file'), (list, tuple)):
dataset = _concat_dataset(cfg, default_args)
else:
dataset = build_from_cfg(cfg, DATASETS, default_args)
return dataset
def build_dataset(cfg, default_args=None):
"""Build a dataset from config dict.
Args:
cfg (dict): Config dict. It should at least contain the key "type".
default_args (dict | None, optional): Default initialization arguments.
Default: None.
Returns:
Dataset: The constructed dataset.
"""
dataset = build_from_cfg(cfg, DATASETS, default_args)
return dataset
def test_resize():
# test assertion if img_scale is a list
with pytest.raises(AssertionError):
transform = dict(type='Resize', img_scale=[1333, 800], keep_ratio=True)
build_from_cfg(transform, PIPELINES)
# test assertion if len(img_scale) while ratio_range is not None
with pytest.raises(AssertionError):
transform = dict(type='Resize',
img_scale=[(1333, 800), (1333, 600)],
ratio_range=(0.9, 1.1),
keep_ratio=True)
build_from_cfg(transform, PIPELINES)
# test assertion for invalid multiscale_mode
with pytest.raises(AssertionError):
transform = dict(type='Resize',
img_scale=[(1333, 800), (1333, 600)],
keep_ratio=True,
multiscale_mode='2333')
build_from_cfg(transform, PIPELINES)
transform = dict(type='Resize', img_scale=(1333, 800), keep_ratio=True)
resize_module = build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(osp.join(osp.dirname(__file__), '../data/color.jpg'),
'color')
results['img'] = img
results['img2'] = copy.deepcopy(img)
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
# Set initial values for default meta_keys
results['pad_shape'] = img.shape
results['scale_factor'] = 1.0
results['img_fields'] = ['img', 'img2']
results = resize_module(results)
assert np.equal(results['img'], results['img2']).all()
results.pop('scale')
transform = dict(type='Resize',
img_scale=(1280, 800),
multiscale_mode='value',
keep_ratio=False)
resize_module = build_from_cfg(transform, PIPELINES)
results = resize_module(results)
assert np.equal(results['img'], results['img2']).all()
assert results['img_shape'] == (800, 1280, 3)
def test_seq_color_jitter(self):
results = self.results.copy()
transform = dict(type='SeqPhotoMetricDistortion', share_params=False)
transform = build_from_cfg(transform, PIPELINES)
outs = transform(results)
assert outs[0]['img_info']['color_jitter'] != outs[1]['img_info'][
'color_jitter']
transform.share_params = True
outs = transform(results)
assert outs[0]['img_info']['color_jitter'] == outs[1]['img_info'][
'color_jitter']
def test_seq_pad(self):
results = copy.deepcopy(self.results)
transform = dict(type='SeqPad', size_divisor=32)
transform = build_from_cfg(transform, PIPELINES)
results = transform(results)
for result in results:
img_shape = result['img'].shape
assert img_shape[0] % 32 == 0
assert img_shape[1] % 32 == 0
resize_transform = dict(type='SeqResize',
img_scale=(1333, 800),
keep_ratio=True)
resize_module = build_from_cfg(resize_transform, PIPELINES)
results = resize_module(results)
results = transform(results)
for result in results:
img_shape = result['img'].shape
assert img_shape[0] % 32 == 0
assert img_shape[1] % 32 == 0
def test_pad():
# test assertion if both size_divisor and size is None
with pytest.raises(AssertionError):
transform = dict(type='Pad')
build_from_cfg(transform, PIPELINES)
transform = dict(type='Pad', size_divisor=32)
transform = build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../../../data/color.jpg'), 'color')
original_img = copy.deepcopy(img)
results['img'] = img
results['img2'] = copy.deepcopy(img)
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
# Set initial values for default meta_keys
results['pad_shape'] = img.shape
results['scale_factor'] = 1.0
results['img_fields'] = ['img', 'img2']
results = transform(results)
assert np.equal(results['img'], results['img2']).all()
# original img already divisible by 32
assert np.equal(results['img'], original_img).all()
img_shape = results['img'].shape
assert img_shape[0] % 32 == 0
assert img_shape[1] % 32 == 0
resize_transform = dict(type='Resize',
img_scale=(1333, 800),
keep_ratio=True)
resize_module = build_from_cfg(resize_transform, PIPELINES)
results = resize_module(results)
results = transform(results)
img_shape = results['img'].shape
assert np.equal(results['img'], results['img2']).all()
assert img_shape[0] % 32 == 0
assert img_shape[1] % 32 == 0
def test_albu_transform():
data_prefix = 'tests/data/coco/'
results = dict(image_file=osp.join(data_prefix, '000000000785.jpg'))
# Define simple pipeline
load = dict(type='LoadImageFromFile')
load = build_from_cfg(load, PIPELINES)
albu_transform = dict(
type='Albumentation',
transforms=[
dict(type='RandomBrightnessContrast', p=0.2),
dict(type='ToFloat')
])
albu_transform = build_from_cfg(albu_transform, PIPELINES)
# Execute transforms
results = load(results)
results = albu_transform(results)
assert results['img'].dtype == np.float32
def test_rotate():
# test assertion degree should be tuple[float] or float
with pytest.raises(AssertionError):
transform = dict(type='RandomRotate', prob=0.5, degree=-10)
build_from_cfg(transform, PIPELINES)
# test assertion degree should be tuple[float] or float
with pytest.raises(AssertionError):
transform = dict(type='RandomRotate', prob=0.5, degree=(10., 20., 30.))
build_from_cfg(transform, PIPELINES)
transform = dict(type='RandomRotate', degree=10., prob=1.)
transform = build_from_cfg(transform, PIPELINES)
assert str(transform) == f'RandomRotate(' \
f'prob={1.}, ' \
f'degree=({-10.}, {10.}), ' \
f'pad_val={0}, ' \
f'seg_pad_val={255}, ' \
f'center={None}, ' \
f'auto_bound={False})'
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../data/color.jpg'), 'color')
h, w, _ = img.shape
seg = np.array(
Image.open(osp.join(osp.dirname(__file__), '../data/seg.png')))
results['img'] = img
results['gt_semantic_seg'] = seg
results['seg_fields'] = ['gt_semantic_seg']
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
# Set initial values for default meta_keys
results['pad_shape'] = img.shape
results['scale_factor'] = 1.0
results = transform(results)
assert results['img'].shape[:2] == (h, w)
assert results['gt_semantic_seg'].shape[:2] == (h, w)
def test_mosaic():
# test assertion for invalid img_scale
with pytest.raises(AssertionError):
transform = dict(type='Mosaic', img_scale=640)
build_from_cfg(transform, PIPELINES)
# test assertion for invalid probability
with pytest.raises(AssertionError):
transform = dict(type='Mosaic', prob=1.5)
build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../../../data/color.jpg'), 'color')
results['img'] = img
# TODO: add img_fields test
results['bbox_fields'] = ['gt_bboxes', 'gt_bboxes_ignore']
h, w, _ = img.shape
gt_bboxes = create_random_bboxes(8, w, h)
gt_bboxes_ignore = create_random_bboxes(2, w, h)
results['gt_labels'] = np.ones(gt_bboxes.shape[0], dtype=np.int64)
results['gt_bboxes'] = gt_bboxes
results['gt_bboxes_ignore'] = gt_bboxes_ignore
transform = dict(type='Mosaic', img_scale=(10, 12))
mosaic_module = build_from_cfg(transform, PIPELINES)
# test assertion for invalid mix_results
with pytest.raises(AssertionError):
mosaic_module(results)
results['mix_results'] = [copy.deepcopy(results)] * 3
results = mosaic_module(results)
assert results['img'].shape[:2] == (20, 24)
assert results['gt_labels'].shape[0] == results['gt_bboxes'].shape[0]
assert results['gt_labels'].dtype == np.int64
assert results['gt_bboxes'].dtype == np.float32
assert results['gt_bboxes_ignore'].dtype == np.float32
def test_random_shift():
# test assertion for invalid shift_ratio
with pytest.raises(AssertionError):
transform = dict(type='RandomShift', shift_ratio=1.5)
build_from_cfg(transform, PIPELINES)
# test assertion for invalid max_shift_px
with pytest.raises(AssertionError):
transform = dict(type='RandomShift', max_shift_px=-1)
build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../../../data/color.jpg'), 'color')
results['img'] = img
# TODO: add img_fields test
results['bbox_fields'] = ['gt_bboxes', 'gt_bboxes_ignore']
def create_random_bboxes(num_bboxes, img_w, img_h):
bboxes_left_top = np.random.uniform(0, 0.5, size=(num_bboxes, 2))
bboxes_right_bottom = np.random.uniform(0.5, 1, size=(num_bboxes, 2))
bboxes = np.concatenate((bboxes_left_top, bboxes_right_bottom), 1)
bboxes = (bboxes * np.array([img_w, img_h, img_w, img_h])).astype(
np.int)
return bboxes
h, w, _ = img.shape
gt_bboxes = create_random_bboxes(8, w, h)
gt_bboxes_ignore = create_random_bboxes(2, w, h)
results['gt_labels'] = torch.ones(gt_bboxes.shape[0])
results['gt_bboxes'] = gt_bboxes
results['gt_bboxes_ignore'] = gt_bboxes_ignore
transform = dict(type='RandomShift', shift_ratio=1.0)
random_shift_module = build_from_cfg(transform, PIPELINES)
results = random_shift_module(results)
assert results['img'].shape[:2] == (h, w)
assert results['gt_labels'].shape[0] == results['gt_bboxes'].shape[0]
def test_solarize_add():
# test assertion for invalid type of magnitude
with pytest.raises(AssertionError):
transform = dict(type='SolarizeAdd', magnitude=(1, 2))
build_from_cfg(transform, PIPELINES)
# test assertion for invalid type of thr
with pytest.raises(AssertionError):
transform = dict(type='SolarizeAdd', magnitude=100, thr=(1, 2))
build_from_cfg(transform, PIPELINES)
# test case when prob=0, therefore no solarize
results = construct_toy_data_photometric()
transform = dict(type='SolarizeAdd', magnitude=100, thr=128, prob=0.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
assert (results['img'] == results['ori_img']).all()
# test case when thr=0, therefore no solarize
results = construct_toy_data_photometric()
transform = dict(type='SolarizeAdd', magnitude=100, thr=0, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
assert (results['img'] == results['ori_img']).all()
# test case when thr=128, magnitude=100
results = construct_toy_data_photometric()
transform = dict(type='SolarizeAdd', magnitude=100, thr=128, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
img_solarized = np.array(
[[100, 128, 255], [101, 227, 254], [102, 129, 253]], dtype=np.uint8)
img_solarized = np.stack([img_solarized, img_solarized, img_solarized],
axis=-1)
assert (results['img'] == img_solarized).all()
assert (results['img'] == results['img2']).all()
# test case when thr=100, magnitude=50
results = construct_toy_data_photometric()
transform = dict(type='SolarizeAdd', magnitude=50, thr=100, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
img_solarized = np.array([[50, 128, 255], [51, 127, 254], [52, 129, 253]],
dtype=np.uint8)
img_solarized = np.stack([img_solarized, img_solarized, img_solarized],
axis=-1)
assert (results['img'] == img_solarized).all()
assert (results['img'] == results['img2']).all()
def test_posterize():
# test assertion for invalid type of bits
with pytest.raises(AssertionError):
transform = dict(type='Posterize', bits=4.5)
build_from_cfg(transform, PIPELINES)
# test assertion for invalid value of bits
with pytest.raises(AssertionError):
transform = dict(type='Posterize', bits=10)
build_from_cfg(transform, PIPELINES)
# test case when prob=0, therefore no posterize
results = construct_toy_data_photometric()
transform = dict(type='Posterize', bits=4, prob=0.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
assert (results['img'] == results['ori_img']).all()
# test case when bits=8, therefore no solarize
results = construct_toy_data_photometric()
transform = dict(type='Posterize', bits=8, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
assert (results['img'] == results['ori_img']).all()
# test case when bits=1
results = construct_toy_data_photometric()
transform = dict(type='Posterize', bits=1, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
img_posterized = np.array([[0, 128, 128], [0, 0, 128], [0, 128, 128]],
dtype=np.uint8)
img_posterized = np.stack([img_posterized, img_posterized, img_posterized],
axis=-1)
assert (results['img'] == img_posterized).all()
assert (results['img'] == results['img2']).all()
# test case when bits=3
results = construct_toy_data_photometric()
transform = dict(type='Posterize', bits=3, prob=1.)
pipeline = build_from_cfg(transform, PIPELINES)
results = pipeline(results)
img_posterized = np.array([[0, 128, 224], [0, 96, 224], [0, 128, 224]],
dtype=np.uint8)
img_posterized = np.stack([img_posterized, img_posterized, img_posterized],
axis=-1)
assert (results['img'] == img_posterized).all()
assert (results['img'] == results['img2']).all()
def test_rename_keys():
results = dict(joints_3d=np.ones([17, 3]),
joints_3d_visible=np.ones([17, 3]))
pipeline = dict(type='RenameKeys',
key_pairs=[('joints_3d', 'target'),
('joints_3d_visible', 'target_weight')])
pipeline = build_from_cfg(pipeline, PIPELINES)
results = pipeline(results)
assert 'joints_3d' not in results
assert 'joints_3d_visible' not in results
assert 'target' in results
assert 'target_weight' in results
assert results['target'].shape == (17, 3)
assert results['target_weight'].shape == (17, 3)
def test_seq_normalize(self):
results = copy.deepcopy(self.results)
img_norm_cfg = dict(mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
to_rgb=True)
transform = dict(type='SeqNormalize', **img_norm_cfg)
transform = build_from_cfg(transform, PIPELINES)
results = transform(results)
mean = np.array(img_norm_cfg['mean'])
std = np.array(img_norm_cfg['std'])
for i, result in enumerate(results):
converted_img = (self.results[i]['img'][..., ::-1] - mean) / std
assert np.allclose(result['img'], converted_img)
def test_seq_color_aug(self):
results = copy.deepcopy(self.results)
imgs_shape = [result['img'].shape for result in results]
transform = dict(type='SeqColorAug',
prob=[1.0, 1.0],
rgb_var=[[-0.55919361, 0.98062831, -0.41940627],
[1.72091413, 0.19879334, -1.82968581],
[4.64467907, 4.73710203, 4.88324118]])
seq_color_aug = build_from_cfg(transform, PIPELINES)
results = seq_color_aug(results)
assert results[0]['img'].shape == imgs_shape[0]
assert results[1]['img'].shape == imgs_shape[0]
def test_seq_bbox_jitter(self):
results = copy.deepcopy(self.results)
for res in results:
res['gt_bboxes'] = random_boxes(1, 256)
res['bbox_fields'] = ['gt_bboxes']
transform = dict(type='SeqBboxJitter',
center_jitter_factor=[0, 4.5],
scale_jitter_factor=[0, 0.5],
crop_size_factor=[2, 5])
seq_bbox_jitter = build_from_cfg(transform, PIPELINES)
results = seq_bbox_jitter(results)
assert results[0]['jittered_bboxes'].shape == (1, 4)
assert results[1]['jittered_bboxes'].shape == (1, 4)
def test_mosaic():
# test assertion for invalid img_scale
with pytest.raises(AssertionError):
transform = dict(type='Mosaic', img_scale=640)
build_from_cfg(transform, PIPELINES)
results = dict()
img = mmcv.imread(
osp.join(osp.dirname(__file__), '../../../data/color.jpg'), 'color')
results['img'] = img
# TODO: add img_fields test
results['bbox_fields'] = ['gt_bboxes', 'gt_bboxes_ignore']
def create_random_bboxes(num_bboxes, img_w, img_h):
bboxes_left_top = np.random.uniform(0, 0.5, size=(num_bboxes, 2))
bboxes_right_bottom = np.random.uniform(0.5, 1, size=(num_bboxes, 2))
bboxes = np.concatenate((bboxes_left_top, bboxes_right_bottom), 1)
bboxes = (bboxes * np.array([img_w, img_h, img_w, img_h])).astype(
np.int)
return bboxes
h, w, _ = img.shape
gt_bboxes = create_random_bboxes(8, w, h)
gt_bboxes_ignore = create_random_bboxes(2, w, h)
results['gt_labels'] = torch.ones(gt_bboxes.shape[0])
results['gt_bboxes'] = gt_bboxes
results['gt_bboxes_ignore'] = gt_bboxes_ignore
transform = dict(type='Mosaic', img_scale=(10, 12))
mosaic_module = build_from_cfg(transform, PIPELINES)
# test assertion for invalid mix_results
with pytest.raises(AssertionError):
mosaic_module(results)
results['mix_results'] = [copy.deepcopy(results)] * 3
results = mosaic_module(results)
assert results['img'].shape[:2] == (20, 24)
def __call__(self, model):
if hasattr(model, 'module'):
model = model.module
conv1_lr_mult = self.paramwise_cfg.get('conv1_lr_mult', 1.)
params = []
for name, param in model.named_parameters():
param_group = {'params': [param]}
if name.startswith('conv1') and param.requires_grad:
param_group['lr'] = self.base_lr * conv1_lr_mult
params.append(param_group)
optimizer_cfg['params'] = params
return build_from_cfg(optimizer_cfg, OPTIMIZERS)
def build_dataset(cfg, default_args=None):
from .dataset_wrappers import (ConcatDataset, RepeatDataset,
ClassBalancedDataset)
if isinstance(cfg, (list, tuple)):
dataset = ConcatDataset([build_dataset(c, default_args) for c in cfg])
elif cfg['type'] == 'RepeatDataset':
dataset = RepeatDataset(build_dataset(cfg['dataset'], default_args),
cfg['times'])
elif cfg['type'] == 'ClassBalancedDataset':
dataset = ClassBalancedDataset(
build_dataset(cfg['dataset'], default_args), cfg['oversample_thr'])
else:
dataset = build_from_cfg(cfg, DATASETS, default_args)
return dataset
def test_seq_shift_scale_aug(self):
results = copy.deepcopy(self.results)
for res in results:
res['gt_bboxes'] = random_boxes(1, 256).numpy()
res['bbox_fields'] = ['gt_bboxes']
transform = dict(type='SeqShiftScaleAug',
target_size=[127, 255],
shift=[4, 64],
scale=[0.05, 0.18])
seq_shift_scale_aug = build_from_cfg(transform, PIPELINES)
results = seq_shift_scale_aug(results)
assert results[0]['img'].shape == (127, 127, 3)
assert results[1]['img'].shape == (255, 255, 3)
def test_seq_crop_like_siamfc(self):
results = copy.deepcopy(self.results)
for res in results:
res['gt_bboxes'] = random_boxes(1, 256)
res['bbox_fields'] = ['gt_bboxes']
transform = dict(type='SeqCropLikeSiamFC',
context_amount=0.5,
exemplar_size=127,
crop_size=511)
seq_crop_like_siamfc = build_from_cfg(transform, PIPELINES)
results = seq_crop_like_siamfc(results)
assert results[0]['img'].shape == (511, 511, 3)
assert results[1]['img'].shape == (511, 511, 3)
def test_seq_random_crop(self):
# test assertion for invalid random crop
with pytest.raises(AssertionError):
transform = dict(type='SeqRandomCrop',
crop_size=(-1, 0),
share_params=False)
build_from_cfg(transform, PIPELINES)
crop_size = (256, 384)
transform = dict(type='SeqRandomCrop',
crop_size=crop_size,
share_params=False)
crop_module = build_from_cfg(transform, PIPELINES)
results = copy.deepcopy(self.results)
for res in results:
res['gt_bboxes'] = random_boxes(8, 256)
res['gt_labels'] = np.random.randint(8)
res['gt_instance_ids'] = np.random.randint(8)
res['gt_bboxes_ignore'] = random_boxes(2, 256)
outs = crop_module(results)
assert len(outs) == len(results)
for res in results:
assert res['img'].shape[:2] == crop_size
# All bboxes should be reserved after crop
assert res['img_shape'][:2] == crop_size
assert res['gt_bboxes'].shape[0] == 8
assert res['gt_bboxes_ignore'].shape[0] == 2
assert outs[0]['img_info']['crop_offsets'] != outs[1]['img_info'][
'crop_offsets']
crop_module.share_params = True
outs = crop_module(results)
assert outs[0]['img_info']['crop_offsets'] == outs[1]['img_info'][
'crop_offsets']
def test_color_jitter():
# read test image
results = dict()
img = mmcv.imread(osp.join(osp.dirname(__file__), '../../data/color.jpg'),
'color')
original_img = copy.deepcopy(img)
results['img'] = img
results['img2'] = copy.deepcopy(img)
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
results['img_fields'] = ['img', 'img2']
def reset_results(results, original_img):
results['img'] = copy.deepcopy(original_img)
results['img2'] = copy.deepcopy(original_img)
results['img_shape'] = original_img.shape
results['ori_shape'] = original_img.shape
return results
transform = dict(type='ColorJitter',
brightness=0.,
contrast=0.,
saturation=0.)
colorjitter_module = build_from_cfg(transform, PIPELINES)
results = colorjitter_module(results)
assert np.equal(results['img'], original_img).all()
assert np.equal(results['img'], results['img2']).all()
results = reset_results(results, original_img)
transform = dict(type='ColorJitter',
brightness=0.3,
contrast=0.3,
saturation=0.3)
colorjitter_module = build_from_cfg(transform, PIPELINES)
results = colorjitter_module(results)
assert not np.equal(results['img'], original_img).all()
def build_dataset(cfg, default_args=None):
if isinstance(cfg, (list, tuple)):
dataset = ConcatDataset([build_dataset(c, default_args) for c in cfg])
elif cfg["type"] == "RepeatDataset":
dataset = RepeatDataset(build_dataset(cfg["dataset"], default_args), cfg["times"])
elif cfg["type"] == "ClassBalancedDataset":
dataset = ClassBalancedDataset(build_dataset(cfg["dataset"], default_args), cfg["oversample_thr"])
elif cfg["type"] == "SourceBalancedDataset":
dataset = SourceBalancedDataset(build_dataset(cfg["dataset"], default_args), cfg["oversample_thr"])
elif isinstance(cfg.get("ann_file"), (list, tuple)):
dataset = _concat_dataset(cfg, default_args)
else:
dataset = build_from_cfg(cfg, DATASETS, default_args)
return dataset
