def test_invert(self):
tf = T.RandomApply(
transforms=[T.Identity(), T.ToTensor()],
p=1,
)
img_tensor = tf(self.img_pil)
self.assertIsInstance(img_tensor, torch.Tensor)
self.assertIsInstance(tf.invert(img_tensor), Image.Image)
def __call__(self, img):
"""
Args:
img (PIL Image): Image to be flipped.
Returns:
PIL Image: Randomly flipped image.
"""
self._transform = T.Identity()
if flip_coin(self.p):
self._transform = HorizontalFlip()
return self._transform(img)
def __call__(self, img):
"""
Args:
img (PIL Image): Image to be converted to grayscale.
Returns:
PIL Image: Randomly grayscaled image.
"""
self._transform = T.Identity()
if flip_coin(self.p):
num_output_channels = 1 if img.mode == 'L' else 3
self._transform = Grayscale(num_output_channels=num_output_channels)
return self._transform(img)
def __call__(self, img):
"""
Args:
img (Tensor): Tensor image of size (C, H, W) to be erased.
Returns:
img (Tensor): Erased Tensor image.
"""
self._transform = T.Identity()
if flip_coin(self.p):
x, y, h, w, v = self.get_params(img, scale=self.scale, ratio=self.ratio, value=self.value)
self._transform = T.Lambda(
lambd=lambda img: F.erase(img, x, y, h, w, v, self.inplace),
tf_inv=lambda img: img,
repr_str='RandomErasing()'
)
return img
def __call__(self, img):
"""
Args:
img (PIL Image): Image to be Perspectively transformed.
Returns:
PIL Image: Random perspectivley transformed image.
"""
self._transform = T.Identity()
if flip_coin(self.p):
width, height = img.size
startpoints, endpoints = self.get_params(width, height, self.distortion_scale)
self._transform = Perspective(
startpoints=startpoints,
endpoints=endpoints,
interpolation=self.interpolation,
)
return self._transform(img)
def setUp(self) -> None:
super().setUp()
self.tf = T.Identity()
def __call__(self, img):
self._transform = T.Identity()
if flip_coin(self.p):
self._transform = Compose(self.transforms)
return self._transform(img)