def init_mnist_cifar_transforms(n_channels=1, stage='train'):
if n_channels == 1:
norm_mean_std = Normalize((0.1307, ), (0.3081, ))
elif n_channels == 3:
norm_mean_std = Normalize((0.4914, 0.4822, 0.4465),
(0.247, 0.243, 0.261))
else:
raise ValueError("Not support channels of {}".format(n_channels))
train_trf = Compose([
wrap2solt,
slc.Stream([
slt.RandomScale(range_x=(0.9, 1.1), same=False, p=0.5),
slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.PadTransform(pad_to=34),
slt.CropTransform(crop_size=32, crop_mode='r')
]), unpack_solt,
ApplyTransform(norm_mean_std)
])
if stage == 'train':
return train_trf
test_trf = Compose([
wrap2solt,
slt.PadTransform(pad_to=32), unpack_solt,
ApplyTransform(norm_mean_std)
])
return test_trf
def init_mnist_transforms():
train_trf = Compose([
wrap2solt,
slc.Stream([
slt.ResizeTransform(resize_to=(64, 64), interpolation='bilinear'),
slt.RandomScale(range_x=(0.9, 1.1), same=False, p=0.5),
slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
slt.RandomRotate(rotation_range=(-10, 10), p=0.5),
# slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.PadTransform(pad_to=70),
slt.CropTransform(crop_size=64, crop_mode='r'),
slt.ImageAdditiveGaussianNoise(p=1.0)
]),
unpack_solt,
ApplyTransform(Normalize((0.5, ), (0.5, )))
])
test_trf = Compose([
wrap2solt,
slt.ResizeTransform(resize_to=(64, 64), interpolation='bilinear'),
# slt.PadTransform(pad_to=64),
unpack_solt,
ApplyTransform(Normalize((0.5, ), (0.5, ))),
])
return train_trf, test_trf
def init_transforms(nc=1):
if nc == 1:
norm_mean_std = Normalize((0.1307, ), (0.3081, ))
elif nc == 3:
norm_mean_std = Normalize((0.4914, 0.4822, 0.4465),
(0.247, 0.243, 0.261))
else:
raise ValueError("Not support channels of {}".format(nc))
train_trf = Compose([
wrap2solt,
slc.Stream([
# slt.ResizeTransform(resize_to=(32, 32), interpolation='bilinear'),
slt.RandomScale(range_x=(0.95, 1.05), same=False, p=0.5),
# slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
slt.RandomRotate(rotation_range=(-10, 10), p=0.5),
slt.RandomFlip(p=0.5, axis=1),
# slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.RandomTranslate(range_x=3, range_y=3),
# slt.PadTransform(pad_to=34),
# slt.CropTransform(crop_size=32, crop_mode='r'),
# slt.ImageAdditiveGaussianNoise(p=1.0)
]),
unpack_solt,
ApplyTransform(norm_mean_std)
])
test_trf = Compose([
wrap2solt,
slt.ResizeTransform(resize_to=(32, 32), interpolation='bilinear'),
unpack_solt,
ApplyTransform(norm_mean_std)
])
def custom_augment(img):
tr = Compose([
wrap2solt,
slc.Stream([
slt.ResizeTransform(resize_to=(32, 32),
interpolation='bilinear'),
slt.RandomScale(range_x=(0.9, 1.1), same=False, p=0.5),
slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
slt.RandomRotate(rotation_range=(-10, 10), p=0.5),
# slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.PadTransform(pad_to=36),
slt.CropTransform(crop_size=32, crop_mode='r'),
slt.ImageAdditiveGaussianNoise(p=1.0)
]),
unpack_solt,
ApplyTransform(norm_mean_std)
])
img_tr, _ = tr((img, 0))
return img_tr
return train_trf, test_trf, custom_augment
def init_transform_wo_aug(nc=1):
if nc == 1:
norm_mean_std = Normalize((0.5, ), (0.5, ))
elif nc == 3:
norm_mean_std = Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
else:
raise ValueError("Not support channels of {}".format(nc))
trf = Compose([wrap2solt, unpack_solt, ApplyTransform(norm_mean_std)])
return trf
def custom_augment(img):
if len(img.shape) == 3:
imgs = img.expand_dims(img, axis=0)
else:
imgs = img
out_imgs = []
for b in range(img.shape[0]):
img1 = imgs[b, :, :, 0:1].astype(np.uint8)
img2 = imgs[b, :, :, 1:2].astype(np.uint8)
tr = Compose([
wrap2solt,
slc.Stream([
slt.ImageAdditiveGaussianNoise(p=0.5, gain_range=0.3),
slt.RandomRotate(p=1, rotation_range=(-10, 10)),
slt.PadTransform(pad_to=int(STD_SZ[0] * 1.05)),
slt.CropTransform(crop_size=STD_SZ[0], crop_mode='r'),
slt.ImageGammaCorrection(p=0.5, gamma_range=(0.5, 1.5)),
]), unpack_solt,
ApplyTransform(Normalize((0.5, ), (0.5, )))
])
img1, _ = tr((img1, 0))
img2, _ = tr((img2, 0))
out_img = torch.cat((img1, img2), dim=0)
out_imgs.append(out_img)
out_imgs = torch.stack(out_imgs, dim=0)
return out_imgs
def init_mnist_transforms():
return Compose([
wrap2solt,
slt.ResizeTransform(resize_to=(64, 64), interpolation='bilinear'),
unpack_solt,
ApplyTransform(Normalize((0.5, ), (0.5, )))
])
def init_transforms():
train_trf = Compose([
wrap2solt,
slc.Stream([
slt.ImageAdditiveGaussianNoise(p=0.5, gain_range=0.3),
slt.RandomRotate(p=1, rotation_range=(-10, 10)),
slt.PadTransform(pad_to=int(STD_SZ[0] * 1.05)),
slt.CropTransform(crop_size=STD_SZ[0], crop_mode='r'),
slt.ImageGammaCorrection(p=0.5, gamma_range=(0.5, 1.5)),
]),
unpack_solt,
ApplyTransform(Normalize((0.5,), (0.5,)))
])
test_trf = Compose([
wrap2solt,
unpack_solt,
ApplyTransform(Normalize((0.5,), (0.5,)))
])
return {"train": train_trf, "eval": test_trf}
def init_transforms(nc=1):
if nc == 1:
norm_mean_std = Normalize((0.5, ), (0.5, ))
elif nc == 3:
norm_mean_std = Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
else:
raise ValueError("Not support channels of {}".format(nc))
train_trf = Compose([
wrap2solt,
slc.Stream([
slt.ImageAdditiveGaussianNoise(p=0.5, gain_range=0.3),
slt.RandomRotate(p=1, rotation_range=(-10, 10)),
slt.PadTransform(pad_to=int(STD_SZ[0] * 1.05)),
slt.CropTransform(crop_size=STD_SZ[0], crop_mode='r'),
slt.ImageGammaCorrection(p=0.5, gamma_range=(0.5, 1.5)),
]), unpack_solt,
ApplyTransform(norm_mean_std)
])
test_trf = Compose([wrap2solt, unpack_solt, ApplyTransform(norm_mean_std)])
return train_trf, test_trf, custom_augment
def init_mnist_transforms():
norm_mean_std = Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
train_trf = Compose([
wrap2solt,
slt.PadTransform(pad_to=32), unpack_solt,
ApplyTransform(norm_mean_std, [0, 1])
])
test_trf = Compose([
wrap2solt,
slt.PadTransform(pad_to=32), unpack_solt,
ApplyTransform(norm_mean_std, [0, 1])
])
return train_trf, test_trf
def custom_augment(img):
img1 = img[:, :, 0:1].astype(np.uint8)
img2 = img[:, :, 1:2].astype(np.uint8)
tr = Compose([
wrap2solt,
slc.Stream([
slt.ImageAdditiveGaussianNoise(p=0.5, gain_range=0.3),
slt.RandomRotate(p=1, rotation_range=(-10, 10)),
slt.PadTransform(pad_to=int(STD_SZ[0] * 1.05)),
slt.CropTransform(crop_size=STD_SZ[0], crop_mode='r'),
slt.ImageGammaCorrection(p=0.5, gamma_range=(0.5, 1.5)),
]), unpack_solt,
ApplyTransform(Normalize((0.5, ), (0.5, )))
])
img1, _ = tr((img1, 0))
img2, _ = tr((img2, 0))
out_img = torch.cat((img1, img2), dim=0)
return out_img
def init_transforms(nc=1):
if nc == 1:
norm_mean_std = Normalize((0.1307, ), (0.3081, ))
elif nc == 3:
norm_mean_std = Normalize((0.4914, 0.4822, 0.4465),
(0.247, 0.243, 0.261))
else:
raise ValueError("Not support channels of {}".format(nc))
train_trf = Compose([
wrap2solt,
slc.Stream([
slt.ResizeTransform(resize_to=(32, 32), interpolation='bilinear'),
slt.RandomScale(range_x=(0.9, 1.1), same=False, p=0.5),
slt.RandomFlip(axis=1, p=0.5),
# slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
# slt.RandomRotate(rotation_range=(-10, 10), p=0.5),
slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.PadTransform(pad_to=36),
slt.CropTransform(crop_size=32, crop_mode='r'),
slt.ImageAdditiveGaussianNoise(p=1.0)
]),
unpack_solt,
ApplyTransform(norm_mean_std)
])
test_trf = Compose([
wrap2solt,
slt.ResizeTransform(resize_to=(32, 32), interpolation='bilinear'),
unpack_solt,
ApplyTransform(norm_mean_std)
])
def custom_augment(img):
tr = Compose([
wrap2solt,
slc.Stream([
slt.ResizeTransform(resize_to=(32, 32),
interpolation='bilinear'),
slt.RandomScale(range_x=(0.9, 1.1), same=False, p=0.5),
slt.RandomFlip(axis=1, p=0.5),
# slt.RandomShear(range_x=(-0.05, 0.05), p=0.5),
# slt.RandomRotate(rotation_range=(-10, 10), p=0.5),
slt.RandomRotate(rotation_range=(-5, 5), p=0.5),
slt.PadTransform(pad_to=36),
slt.CropTransform(crop_size=32, crop_mode='r'),
slt.ImageAdditiveGaussianNoise(p=1.0)
]),
unpack_solt,
ApplyTransform(norm_mean_std)
])
if len(img.shape) == 3:
imgs = np.expand_dims(img, axis=0)
elif len(img.shape) == 4:
imgs = img
else:
raise ValueError('Expect num of dims 3 or 4, but got {}'.format(
len(img.shape)))
out_imgs = []
for b in range(imgs.shape[0]):
_img = imgs[b, :].astype(np.uint8)
_img, _ = tr((_img, 0))
out_imgs.append(_img)
return torch.stack(out_imgs, dim=0)
return train_trf, test_trf, custom_augment
