Pytorch Transforms Tensor
transforms代码
[docs]class Compose(object):
"""Composes several transforms together.
Args:
transforms (list of ``Transform`` objects): list of transforms to compose.
Example:
>>> transforms.Compose([
>>> transforms.CenterCrop(10),
>>> transforms.ToTensor(),
>>> ])
"""
def __init__(self, transforms):
self.transforms = transforms
def __call__(self, img):
for t in self.transforms:
img = t(img)
return img
def __repr__(self):
format_string = self.__class__.__name__ + '('
for t in self.transforms:
format_string += '\n'
format_string += ' {0}'.format(t)
format_string += '\n)'
return format_stringtransforms.ToTensor()代码，把PIL.Image.Image和numpy.ndarray转换成Tensor，把（H,W,C)转换成（C,H,W)，把（H,W)转换成（1，H,W)，若原数据类型是uint8,则归一化到（0,255）。
[docs]class ToTensor(object):
"""Convert a ``PIL Image`` or ``numpy.ndarray`` to tensor.
Converts a PIL Image or numpy.ndarray (H x W x C) in the range
[0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0]
if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1)
or if the numpy.ndarray has dtype = np.uint8
In the other cases, tensors are returned without scaling.
"""
[docs] def __call__(self, pic):
"""
Args:
pic (PIL Image or numpy.ndarray): Image to be converted to tensor.
Returns:
Tensor: Converted image.
"""
return F.to_tensor(pic)
def __repr__(self):
return self.__class__.__name__ + '()'
def to_tensor(pic):
"""Convert a ``PIL Image`` or ``numpy.ndarray`` to tensor.
See ``ToTensor`` for more details.
Args:
pic (PIL Image or numpy.ndarray): Image to be converted to tensor.
Returns:
Tensor: Converted image.
"""
if not(_is_pil_image(pic) or _is_numpy(pic)):
raise TypeError('pic should be PIL Image or ndarray. Got {}'.format(type(pic)))
if _is_numpy(pic) and not _is_numpy_image(pic):
raise ValueError('pic should be 2/3 dimensional. Got {} dimensions.'.format(pic.ndim))
if isinstance(pic, np.ndarray):
# handle numpy array
if pic.ndim == 2:
pic = pic[:, :, None]
img = torch.from_numpy(pic.transpose((2, 0, 1)))
# backward compatibility
if isinstance(img, torch.ByteTensor):
return img.float().div(255)
else:
return img
if accimage is not None and isinstance(pic, accimage.Image):
nppic = np.zeros([pic.channels, pic.height, pic.width], dtype=np.float32)
pic.copyto(nppic)
return torch.from_numpy(nppic)
# handle PIL Image
if pic.mode == 'I':
img = torch.from_numpy(np.array(pic, np.int32, copy=False))
elif pic.mode == 'I;16':
img = torch.from_numpy(np.array(pic, np.int16, copy=False))
elif pic.mode == 'F':
img = torch.from_numpy(np.array(pic, np.float32, copy=False))
elif pic.mode == '1':
img = 255 * torch.from_numpy(np.array(pic, np.uint8, copy=False))
else:
img = torch.ByteTensor(torch.ByteStorage.from_buffer(pic.tobytes()))
# PIL image mode: L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK
if pic.mode == 'YCbCr':
nchannel = 3
elif pic.mode == 'I;16':
nchannel = 1
else:
nchannel = len(pic.mode)
img = img.view(pic.size[1], pic.size[0], nchannel)
# put it from HWC to CHW format
# yikes, this transpose takes 80% of the loading time/CPU
img = img.transpose(0, 1).transpose(0, 2).contiguous()
if isinstance(img, torch.ByteTensor):
return img.float().div(255)
else:
return img将ndarray或tensor转换为PIL.Image。mode为Image的数据类型，可以为空。
[docs]class ToPILImage(object):
"""Convert a tensor or an ndarray to PIL Image.
Converts a torch.*Tensor of shape C x H x W or a numpy ndarray of shape
H x W x C to a PIL Image while preserving the value range.
Args:
mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).
If ``mode`` is ``None`` (default) there are some assumptions made about the input data:
- If the input has 4 channels, the ``mode`` is assumed to be ``RGBA``.
- If the input has 3 channels, the ``mode`` is assumed to be ``RGB``.
- If the input has 2 channels, the ``mode`` is assumed to be ``LA``.
- If the input has 1 channel, the ``mode`` is determined by the data type (i.e ``int``, ``float``,
``short``).
.. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes
"""
def __init__(self, mode=None):
self.mode = mode
[docs] def __call__(self, pic):
"""
Args:
pic (Tensor or numpy.ndarray): Image to be converted to PIL Image.
Returns:
PIL Image: Image converted to PIL Image.
"""
return F.to_pil_image(pic, self.mode)
def __repr__(self):
format_string = self.__class__.__name__ + '('
if self.mode is not None:
format_string += 'mode={0}'.format(self.mode)
format_string += ')'
return format_string
def to_pil_image(pic, mode=None):
"""Convert a tensor or an ndarray to PIL Image.
See :class:`~torchvision.transforms.ToPILImage` for more details.
Args:
pic (Tensor or numpy.ndarray): Image to be converted to PIL Image.
mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).
.. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes
Returns:
PIL Image: Image converted to PIL Image.
"""
if not(isinstance(pic, torch.Tensor) or isinstance(pic, np.ndarray)):
raise TypeError('pic should be Tensor or ndarray. Got {}.'.format(type(pic)))
elif isinstance(pic, torch.Tensor):
if pic.ndimension() not in {2, 3}:
raise ValueError('pic should be 2/3 dimensional. Got {} dimensions.'.format(pic.ndimension()))
elif pic.ndimension() == 2:
# if 2D image, add channel dimension (CHW)
pic = pic.unsqueeze(0)
elif isinstance(pic, np.ndarray):
if pic.ndim not in {2, 3}:
raise ValueError('pic should be 2/3 dimensional. Got {} dimensions.'.format(pic.ndim))
elif pic.ndim == 2:
# if 2D image, add channel dimension (HWC)
pic = np.expand_dims(pic, 2)
npimg = pic
if isinstance(pic, torch.FloatTensor) and mode != 'F':
pic = pic.mul(255).byte()
if isinstance(pic, torch.Tensor):
npimg = np.transpose(pic.numpy(), (1, 2, 0))
if not isinstance(npimg, np.ndarray):
raise TypeError('Input pic must be a torch.Tensor or NumPy ndarray, ' +
'not {}'.format(type(npimg)))
if npimg.shape[2] == 1:
expected_mode = None
npimg = npimg[:, :, 0]
if npimg.dtype == np.uint8:
expected_mode = 'L'
elif npimg.dtype == np.int16:
expected_mode = 'I;16'
elif npimg.dtype == np.int32:
expected_mode = 'I'
elif npimg.dtype == np.float32:
expected_mode = 'F'
if mode is not None and mode != expected_mode:
raise ValueError("Incorrect mode ({}) supplied for input type {}. Should be {}"
.format(mode, np.dtype, expected_mode))
mode = expected_mode
elif npimg.shape[2] == 2:
permitted_2_channel_modes = ['LA']
if mode is not None and mode not in permitted_2_channel_modes:
raise ValueError("Only modes {} are supported for 2D inputs".format(permitted_2_channel_modes))
if mode is None and npimg.dtype == np.uint8:
mode = 'LA'
elif npimg.shape[2] == 4:
permitted_4_channel_modes = ['RGBA', 'CMYK', 'RGBX']
if mode is not None and mode not in permitted_4_channel_modes:
raise ValueError("Only modes {} are supported for 4D inputs".format(permitted_4_channel_modes))
if mode is None and npimg.dtype == np.uint8:
mode = 'RGBA'
else:
permitted_3_channel_modes = ['RGB', 'YCbCr', 'HSV']
if mode is not None and mode not in permitted_3_channel_modes:
raise ValueError("Only modes {} are supported for 3D inputs".format(permitted_3_channel_modes))
if mode is None and npimg.dtype == np.uint8:
mode = 'RGB'
if mode is None:
raise TypeError('Input type {} is not supported'.format(npimg.dtype))
return Image.fromarray(npimg, mode=mode)transforms.Normalize，创建Normalize对象是大小为C的mean和std列表，调用是（C，H，W）的tensor_image。
[docs]class Normalize(object):
"""Normalize a tensor image with mean and standard deviation.
Given mean: ``(M1,...,Mn)`` and std: ``(S1,..,Sn)`` for ``n`` channels, this transform
will normalize each channel of the input ``torch.*Tensor`` i.e.
``input[channel] = (input[channel] - mean[channel]) / std[channel]``
.. note::
This transform acts out of place, i.e., it does not mutates the input tensor.
Args:
mean (sequence): Sequence of means for each channel.
std (sequence): Sequence of standard deviations for each channel.
inplace(bool,optional): Bool to make this operation in-place.
"""
def __init__(self, mean, std, inplace=False):
self.mean = mean
self.std = std
self.inplace = inplace
[docs] def __call__(self, tensor):
"""
Args:
tensor (Tensor): Tensor image of size (C, H, W) to be normalized.
Returns:
Tensor: Normalized Tensor image.
"""
return F.normalize(tensor, self.mean, self.std, self.inplace)
def __repr__(self):
return self.__class__.__name__ + '(mean={0}, std={1})'.format(self.mean, self.std)
def normalize(tensor, mean, std, inplace=False):
"""Normalize a tensor image with mean and standard deviation.
.. note::
This transform acts out of place by default, i.e., it does not mutates the input tensor.
See :class:`~torchvision.transforms.Normalize` for more details.
Args:
tensor (Tensor): Tensor image of size (C, H, W) to be normalized.
mean (sequence): Sequence of means for each channel.
std (sequence): Sequence of standard deviations for each channel.
inplace(bool,optional): Bool to make this operation inplace.
Returns:
Tensor: Normalized Tensor image.
"""
if not _is_tensor_image(tensor):
raise TypeError('tensor is not a torch image.')
if not inplace:
tensor = tensor.clone()
dtype = tensor.dtype
mean = torch.as_tensor(mean, dtype=dtype, device=tensor.device)
std = torch.as_tensor(std, dtype=dtype, device=tensor.device)
#将mean和std转换成（C，1,1)
tensor.sub_(mean[:, None, None]).div_(std[:, None, None])
return tensortransform.LinearTransformation。先把图像flatten成（1，C*H*W)的形状的向量，再减去mean_vector (Tensor): tensor （1， D = C x H x W）乘以transformation_matrix (Tensor): tensor （D = C x H x W，D = C x H x W），再flat_tensor.view(tensor.size())。
[docs]class LinearTransformation(object):
"""Transform a tensor image with a square transformation matrix and a mean_vector computed
offline.
Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and
subtract mean_vector from it which is then followed by computing the dot
product with the transformation matrix and then reshaping the tensor to its
original shape.
Applications:
whitening transformation: Suppose X is a column vector zero-centered data.
Then compute the data covariance matrix [D x D] with torch.mm(X.t(), X),
perform SVD on this matrix and pass it as transformation_matrix.
Args:
transformation_matrix (Tensor): tensor [D x D], D = C x H x W
mean_vector (Tensor): tensor [D], D = C x H x W
"""
def __init__(self, transformation_matrix, mean_vector):
if transformation_matrix.size(0) != transformation_matrix.size(1):
raise ValueError("transformation_matrix should be square. Got " +
"[{} x {}] rectangular matrix.".format(*transformation_matrix.size()))
if mean_vector.size(0) != transformation_matrix.size(0):
raise ValueError("mean_vector should have the same length {}".format(mean_vector.size(0)) +
" as any one of the dimensions of the transformation_matrix [{} x {}]"
.format(transformation_matrix.size()))
self.transformation_matrix = transformation_matrix
self.mean_vector = mean_vector
def __call__(self, tensor):
"""
Args:
tensor (Tensor): Tensor image of size (C, H, W) to be whitened.
Returns:
Tensor: Transformed image.
"""
if tensor.size(0) * tensor.size(1) * tensor.size(2) != self.transformation_matrix.size(0):
raise ValueError("tensor and transformation matrix have incompatible shape." +
"[{} x {} x {}] != ".format(*tensor.size()) +
"{}".format(self.transformation_matrix.size(0)))
flat_tensor = tensor.view(1, -1) - self.mean_vector
transformed_tensor = torch.mm(flat_tensor, self.transformation_matrix)
tensor = transformed_tensor.view(tensor.size())
return tensor
def __repr__(self):
format_string = self.__class__.__name__ + '(transformation_matrix='
format_string += (str(self.transformation_matrix.tolist()) + ')')
format_string += (", (mean_vector=" + str(self.mean_vector.tolist()) + ')')
return format_stringtransforms.RandomErasing。输入torch_image，（C，H，W），随机是否擦除，随机擦除面积，随机擦除长宽比例，随机位置，随机通道，随机值擦除。
randomerase=transforms.RandomErasing(p=1,scale=(0.2,0.5),value=(255,255,0))
im=Image.open(r'C:\Users\Administrator\Desktop\panda.jpg')
#im_data=torch.from_numpy(np.array(im))
im_data_t=transforms.ToTensor()(im)
erased=randomerase(im_data_t)
erased_im=transforms.ToPILImage()(erased)
[docs]class RandomErasing(object):
""" Randomly selects a rectangle region in an image and erases its pixels.
'Random Erasing Data Augmentation' by Zhong et al.
See https://arxiv.org/pdf/1708.04896.pdf
Args:
p: probability that the random erasing operation will be performed.
scale: range of proportion of erased area against input image.
ratio: range of aspect ratio of erased area.
value: erasing value. Default is 0. If a single int, it is used to
erase all pixels. If a tuple of length 3, it is used to erase
R, G, B channels respectively.
If a str of 'random', erasing each pixel with random values.
inplace: boolean to make this transform inplace. Default set to False.
Returns:
Erased Image.
# Examples:
>>> transform = transforms.Compose([
>>> transforms.RandomHorizontalFlip(),
>>> transforms.ToTensor(),
>>> transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),
>>> transforms.RandomErasing(),
>>> ])
"""
def __init__(self, p=0.5, scale=(0.02, 0.33), ratio=(0.3, 3.3), value=0, inplace=False):
assert isinstance(value, (numbers.Number, str, tuple, list))
if (scale[0] > scale[1]) or (ratio[0] > ratio[1]):
warnings.warn("range should be of kind (min, max)")
if scale[0] < 0 or scale[1] > 1:
raise ValueError("range of scale should be between 0 and 1")
if p < 0 or p > 1:
raise ValueError("range of random erasing probability should be between 0 and 1")
self.p = p
self.scale = scale
self.ratio = ratio
self.value = value
self.inplace = inplace
@staticmethod
def get_params(img, scale, ratio, value=0):
"""Get parameters for ``erase`` for a random erasing.
Args:
img (Tensor): Tensor image of size (C, H, W) to be erased.
scale: range of proportion of erased area against input image.
ratio: range of aspect ratio of erased area.
Returns:
tuple: params (i, j, h, w, v) to be passed to ``erase`` for random erasing.
"""
img_c, img_h, img_w = img.shape
area = img_h * img_w
for attempt in range(10):
erase_area = random.uniform(scale[0], scale[1]) * area
aspect_ratio = random.uniform(ratio[0], ratio[1])
h = int(round(math.sqrt(erase_area * aspect_ratio)))
w = int(round(math.sqrt(erase_area / aspect_ratio)))
if h < img_h and w < img_w:
i = random.randint(0, img_h - h)
j = random.randint(0, img_w - w)
if isinstance(value, numbers.Number):
v = value
elif isinstance(value, torch._six.string_classes):
v = torch.empty([img_c, h, w], dtype=torch.float32).normal_()
elif isinstance(value, (list, tuple)):
#转换成相同维度，和矩形框相同大小
v = torch.tensor(value, dtype=torch.float32).view(-1, 1, 1).expand(-1, h, w)
return i, j, h, w, v
# Return original image
return 0, 0, img_h, img_w, img
def __call__(self, img):
"""
Args:
img (Tensor): Tensor image of size (C, H, W) to be erased.
Returns:
img (Tensor): Erased Tensor image.
"""
if random.uniform(0, 1) < self.p:
x, y, h, w, v = self.get_params(img, scale=self.scale, ratio=self.ratio, value=self.value)
return F.erase(img, x, y, h, w, v, self.inplace)
return img
def erase(img, i, j, h, w, v, inplace=False):
""" Erase the input Tensor Image with given value.
Args:
img (Tensor Image): Tensor image of size (C, H, W) to be erased
i (int): i in (i,j) i.e coordinates of the upper left corner.
j (int): j in (i,j) i.e coordinates of the upper left corner.
h (int): Height of the erased region.
w (int): Width of the erased region.
v: Erasing value.
inplace(bool, optional): For in-place operations. By default is set False.
Returns:
Tensor Image: Erased image.
"""
if not isinstance(img, torch.Tensor):
raise TypeError('img should be Tensor Image. Got {}'.format(type(img)))
if not inplace:
img = img.clone()
img[:, i:i + h, j:j + w] = v
return img