原文地址：https://blog.csdn.net/u014380165/article/details/79167753

PyTorch框架中有一個非常重要且好用的包：torchvision，該包主要由3個子包組成，分別是：torchvision.datasets、torchvision.models、torchvision.transforms。這3個子包的具體介紹可以參考官網：http://pytorch.org/docs/master/torchvision/index.html。具體代碼可以參考github：https://github.com/pytorch/vision/tree/master/torchvision。

這篇博客介紹torchvision.transformas。torchvision.transforms這個包中包含resize、crop等常見的data augmentation操作，基本上PyTorch中的data augmentation操作都可以通過該接口實現。該包主要包含兩個腳本：transformas.py和functional.py，前者定義了各種data augmentation的類，在每個類中通過調用functional.py中對應的函數完成data augmentation操作。

使用例子：

import torchvision
import torch
train_augmentation = torchvision.transforms.Compose([torchvision.transforms.Resize(256
 
),
                                                    torchvision.transforms.RandomCrop(224),                                                                            
                                                    torchvision.transofrms.RandomHorizontalFlip(),
                                                    torchvision.transforms.ToTensor(),
                                                    torch vision.Normalize([0.485
 
, 0.456, -.406],[0.229, 0.224, 0.225])
                                                    ])

Class custom_dataread(torch.utils.data.Dataset):
    def __init__():
        ...
    def __getitem__():
        # use self.transform for input image
    def __len__():
        ...

train_loader = torch.utils.data.DataLoader(
    custom_dataread(transform=train_augmentation),
    batch_size = batch_size, shuffle = True,
    num_workers = workers, pin_memory = True)

這裏定義了resize、crop、normalize等數據預處理操作，並最終作為數據讀取類custom_dataread的一個參數傳入，可以在內部方法__getitem__中實現數據增強操作。

主要代碼在transformas.py腳本中，這裏僅介紹常見的data augmentation操作，源碼如下：
首先是導入必須的模型，這裏比較重要的是from . import functional as F，也就是導入了functional.py腳本中具體的data augmentation函數。__all__列表定義了可以從外部import的函數名或類名。

from __future__ import division
import torch
import math
import random
from PIL import Image, ImageOps, ImageEnhance
try:
    import accimage
except ImportError:
    accimage = None
import numpy as np
import numbers
import types
import collections
import warnings

from . import functional as F

__all__ = ["Compose", "ToTensor", "ToPILImage", "Normalize", "Resize",
"Scale", "CenterCrop", "Pad", "Lambda", "RandomCrop", 
"RandomHorizontalFlip", "RandomVerticalFlip", "RandomResizedCrop", 
"RandomSizedCrop", "FiveCrop", "TenCrop","LinearTransformation", 
"ColorJitter", "RandomRotation", "Grayscale", "RandomGrayscale"]

Compose這個類是用來管理各個transform的，可以看到主要的__call__方法就是對輸入圖像img循環所有的transform操作

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_string

ToTensor類是實現：Convert a PIL Image or numpy.ndarray to tensor 的過程，在PyTorch中常用PIL庫來讀取圖像數據，因此這個方法相當於搭建了PIL Image和Tensor的橋梁。另外要強調的是在做數據歸一化之前必須要把PIL Image轉成Tensor，而其他resize或crop操作則不需要。

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].
    """

    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__ + ‘()‘

ToPILImage顧名思義是從Tensor到PIL Image的過程，和前面ToTensor類的相反的操作。

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:
            1. If the input has 3 channels, the ``mode`` is assumed to be ``RGB``.
            2. If the input has 4 channels, the ``mode`` is assumed to be ``RGBA``.
            3. If the input has 1 channel, the ``mode`` is determined by the data type (i,e,
            ``int``, ``float``, ``short``).

    .. _PIL.Image mode: http://pillow.readthedocs.io/en/3.4.x/handbook/concepts.html#modes
    """
    def __init__(self, mode=None):
        self.mode = mode

    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):
        return self.__class__.__name__ + ‘({0})‘.format(self.mode)

Normalize類是做數據歸一化的，一般都會對輸入數據做這樣的操作，公式也在註釋中給出了，比較容易理解。前面提到在調用Normalize的時候，輸入得是Tensor，這個從__call__方法的輸入也可以看出來了。

class Normalize(object):
    """Normalize an 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]``

    Args:
        mean (sequence): Sequence of means for each channel.
        std (sequence): Sequence of standard deviations for each channel.
    """

    def __init__(self, mean, std):
        self.mean = mean
        self.std = std

    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)

    def __repr__(self):
        return self.__class__.__name__ + ‘(mean={0}, std={1})‘.format(self.mean, self.std)

Resize類是對PIL Image做resize操作的，幾乎都要用到。這裏輸入可以是int，此時表示將輸入圖像的短邊resize到這個int數，長邊則根據對應比例調整，圖像的長寬比不變。如果輸入是個(h,w)的序列，h和w都是int，則直接將輸入圖像resize到這個(h,w)尺寸，相當於force resize，所以一般最後圖像的長寬比會變化，也就是圖像內容被拉長或縮短。註意，在__call__方法中調用了functional.py腳本中的resize函數來完成resize操作，因為輸入是PIL Image，所以resize函數基本是在調用Image的各種方法。如果輸入是Tensor，則對應函數基本是在調用Tensor的各種方法，這就是functional.py中的主要內容。

class Resize(object):
    """Resize the input PIL Image to the given size.

    Args:
        size (sequence or int): Desired output size. If size is a sequence like
            (h, w), output size will be matched to this. If size is an int,
            smaller edge of the image will be matched to this number.
            i.e, if height > width, then image will be rescaled to
            (size * height / width, size)
        interpolation (int, optional): Desired interpolation. Default is
            ``PIL.Image.BILINEAR``
    """

    def __init__(self, size, interpolation=Image.BILINEAR):
        assert isinstance(size, int) or (isinstance(size, collections.Iterable) and len(size) == 2)
        self.size = size
        self.interpolation = interpolation

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be scaled.

        Returns:
            PIL Image: Rescaled image.
        """
        return F.resize(img, self.size, self.interpolation)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

CenterCrop是以輸入圖的中心點為中心點做指定size的crop操作，一般數據增強不會采用這個，因為當size固定的時候，在相同輸入圖像的情況下，N次CenterCrop的結果都是一樣的。註釋裏面說明了size為int和序列時候尺寸的定義。

class CenterCrop(object):
    """Crops the given PIL Image at the center.

    Args:
        size (sequence or int): Desired output size of the crop. If size is an
            int instead of sequence like (h, w), a square crop (size, size) is
            made.
    """

    def __init__(self, size):
        if isinstance(size, numbers.Number):
            self.size = (int(size), int(size))
        else:
            self.size = size

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be cropped.

        Returns:
            PIL Image: Cropped image.
        """
        return F.center_crop(img, self.size)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

相比前面的CenterCrop，這個RandomCrop更常用，差別就在於crop時的中心點坐標是隨機的，並不是輸入圖像的中心點坐標，因此基本上每次crop生成的圖像都是有差異的。就是通過 i = random.randint(0, h - th)和 j = random.randint(0, w - tw)兩行生成一個隨機中心點的橫縱坐標。註意到在__call__中最後是調用了F.crop(img, i, j, h, w)來完成crop操作，其實前面CenterCrop中雖然是調用 F.center_crop(img, self.size)，但是在F.center_crop()函數中只是先計算了中心點坐標，最後還是調用F.crop(img, i, j, h, w)完成crop操作。

class RandomCrop(object):
    """Crop the given PIL Image at a random location.

    Args:
        size (sequence or int): Desired output size of the crop. If size is an
            int instead of sequence like (h, w), a square crop (size, size) is
            made.
        padding (int or sequence, optional): Optional padding on each border
            of the image. Default is 0, i.e no padding. If a sequence of length
            4 is provided, it is used to pad left, top, right, bottom borders
            respectively.
    """

    def __init__(self, size, padding=0):
        if isinstance(size, numbers.Number):
            self.size = (int(size), int(size))
        else:
            self.size = size
        self.padding = padding

    @staticmethod
    def get_params(img, output_size):
        """Get parameters for ``crop`` for a random crop.

        Args:
            img (PIL Image): Image to be cropped.
            output_size (tuple): Expected output size of the crop.

        Returns:
            tuple: params (i, j, h, w) to be passed to ``crop`` for random crop.
        """
        w, h = img.size
        th, tw = output_size
        if w == tw and h == th:
            return 0, 0, h, w

        i = random.randint(0, h - th)
        j = random.randint(0, w - tw)
        return i, j, th, tw

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be cropped.

        Returns:
            PIL Image: Cropped image.
        """
        if self.padding > 0:
            img = F.pad(img, self.padding)

        i, j, h, w = self.get_params(img, self.size)

        return F.crop(img, i, j, h, w)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

RandomHorizontalFlip類也是比較常用的，是隨機的圖像水平翻轉，通俗講就是圖像的左右對調。從該類中的__call__方法可以看出水平翻轉的概率是0.5。

class RandomHorizontalFlip(object):
    """Horizontally flip the given PIL Image randomly with a probability of 0.5."""

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be flipped.

        Returns:
            PIL Image: Randomly flipped image.
        """
        if random.random() < 0.5:
            return F.hflip(img)
        return img

    def __repr__(self):
        return self.__class__.__name__ + ‘()‘

同樣的，RandomVerticalFlip類是隨機的圖像豎直翻轉，通俗講就是圖像的上下對調。

class RandomVerticalFlip(object):
    """Vertically flip the given PIL Image randomly with a probability of 0.5."""

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be flipped.

        Returns:
            PIL Image: Randomly flipped image.
        """
        if random.random() < 0.5:
            return F.vflip(img)
        return img

    def __repr__(self):
        return self.__class__.__name__ + ‘()‘

RandomResizedCrop類也是比較常用的，個人非常喜歡用。前面不管是CenterCrop還是RandomCrop，在crop的時候其尺寸是固定的，而這個類則是random size的crop。該類主要用到3個參數：size、scale和ratio，總的來講就是先做crop（用到scale和ratio），再resize到指定尺寸（用到size）。做crop的時候，其中心點坐標和長寬是由get_params方法得到的，在get_params方法中主要用到兩個參數：scale和ratio，首先在scale限定的數值範圍內隨機生成一個數，用這個數乘以輸入圖像的面積作為crop後圖像的面積；然後在ratio限定的數值範圍內隨機生成一個數，表示長寬的比值，根據這兩個值就可以得到crop圖像的長寬了。至於crop圖像的中心點坐標，也是類似RandomCrop類一樣是隨機生成的。

class RandomResizedCrop(object):
    """Crop the given PIL Image to random size and aspect ratio.

    A crop of random size (default: of 0.08 to 1.0) of the original size and a random
    aspect ratio (default: of 3/4 to 4/3) of the original aspect ratio is made. This crop
    is finally resized to given size.
    This is popularly used to train the Inception networks.

    Args:
        size: expected output size of each edge
        scale: range of size of the origin size cropped
        ratio: range of aspect ratio of the origin aspect ratio cropped
        interpolation: Default: PIL.Image.BILINEAR
    """

    def __init__(self, size, scale=(0.08, 1.0), ratio=(3. / 4., 4. / 3.), interpolation=Image.BILINEAR):
        self.size = (size, size)
        self.interpolation = interpolation
        self.scale = scale
        self.ratio = ratio

    @staticmethod
    def get_params(img, scale, ratio):
        """Get parameters for ``crop`` for a random sized crop.

        Args:
            img (PIL Image): Image to be cropped.
            scale (tuple): range of size of the origin size cropped
            ratio (tuple): range of aspect ratio of the origin aspect ratio cropped

        Returns:
            tuple: params (i, j, h, w) to be passed to ``crop`` for a random
                sized crop.
        """
        for attempt in range(10):
            area = img.size[0] * img.size[1]
            target_area = random.uniform(*scale) * area
            aspect_ratio = random.uniform(*ratio)

            w = int(round(math.sqrt(target_area * aspect_ratio)))
            h = int(round(math.sqrt(target_area / aspect_ratio)))

            if random.random() < 0.5:
                w, h = h, w

            if w <= img.size[0] and h <= img.size[1]:
                i = random.randint(0, img.size[1] - h)
                j = random.randint(0, img.size[0] - w)
                return i, j, h, w

        # Fallback
        w = min(img.size[0], img.size[1])
        i = (img.size[1] - w) // 2
        j = (img.size[0] - w) // 2
        return i, j, w, w

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be flipped.

        Returns:
            PIL Image: Randomly cropped and resize image.
        """
        i, j, h, w = self.get_params(img, self.scale, self.ratio)
        return F.resized_crop(img, i, j, h, w, self.size, self.interpolation)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

FiveCrop類，顧名思義就是從一張輸入圖像中crop出5張指定size的圖像，這5張圖像包括4個角的圖像和一個center crop的圖像。曾在TSN算法的看到過這種用法。

class FiveCrop(object):
    """Crop the given PIL Image into four corners and the central crop

    .. Note::
         This transform returns a tuple of images and there may be a mismatch in the number of
         inputs and targets your Dataset returns. See below for an example of how to deal with
         this.

    Args:
         size (sequence or int): Desired output size of the crop. If size is an ``int``
            instead of sequence like (h, w), a square crop of size (size, size) is made.

    Example:
         >>> transform = Compose([
         >>>    FiveCrop(size), # this is a list of PIL Images
         >>>    Lambda(lambda crops: torch.stack([ToTensor()(crop) for crop in crops])) # returns a 4D tensor
         >>> ])
         >>> #In your test loop you can do the following:
         >>> input, target = batch # input is a 5d tensor, target is 2d
         >>> bs, ncrops, c, h, w = input.size()
         >>> result = model(input.view(-1, c, h, w)) # fuse batch size and ncrops
         >>> result_avg = result.view(bs, ncrops, -1).mean(1) # avg over crops
    """

    def __init__(self, size):
        self.size = size
        if isinstance(size, numbers.Number):
            self.size = (int(size), int(size))
        else:
            assert len(size) == 2, "Please provide only two dimensions (h, w) for size."
            self.size = size

    def __call__(self, img):
        return F.five_crop(img, self.size)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

TenCrop類和前面FiveCrop類類似，只不過在FiveCrop的基礎上，再將輸入圖像進行水平或豎直翻轉，然後再進行FiveCrop操作，這樣一張輸入圖像就能得到10張crop結果。

class TenCrop(object):
    """Crop the given PIL Image into four corners and the central crop plus the flipped version of
    these (horizontal flipping is used by default)

    .. Note::
         This transform returns a tuple of images and there may be a mismatch in the number of
         inputs and targets your Dataset returns. See below for an example of how to deal with
         this.

    Args:
        size (sequence or int): Desired output size of the crop. If size is an
            int instead of sequence like (h, w), a square crop (size, size) is
            made.
        vertical_flip(bool): Use vertical flipping instead of horizontal

    Example:
         >>> transform = Compose([
         >>>    TenCrop(size), # this is a list of PIL Images
         >>>    Lambda(lambda crops: torch.stack([ToTensor()(crop) for crop in crops])) # returns a 4D tensor
         >>> ])
         >>> #In your test loop you can do the following:
         >>> input, target = batch # input is a 5d tensor, target is 2d
         >>> bs, ncrops, c, h, w = input.size()
         >>> result = model(input.view(-1, c, h, w)) # fuse batch size and ncrops
         >>> result_avg = result.view(bs, ncrops, -1).mean(1) # avg over crops
    """

    def __init__(self, size, vertical_flip=False):
        self.size = size
        if isinstance(size, numbers.Number):
            self.size = (int(size), int(size))
        else:
            assert len(size) == 2, "Please provide only two dimensions (h, w) for size."
            self.size = size
        self.vertical_flip = vertical_flip

    def __call__(self, img):
        return F.ten_crop(img, self.size, self.vertical_flip)

    def __repr__(self):
        return self.__class__.__name__ + ‘(size={0})‘.format(self.size)

LinearTransformation類是用一個變換矩陣去乘輸入圖像得到輸出結果。

class LinearTransformation(object):
    """Transform a tensor image with a square transformation matrix computed
    offline.

    Given transformation_matrix, will flatten the torch.*Tensor, compute the dot
    product with the transformation matrix and reshape the tensor to its
    original shape.

    Applications:
    - whitening: zero-center the data, compute the data covariance matrix
                 [D x D] with np.dot(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
    """

    def __init__(self, transformation_matrix):
        if transformation_matrix.size(0) != transformation_matrix.size(1):
            raise ValueError("transformation_matrix should be square. Got " +
                             "[{} x {}] rectangular matrix.".format(*transformation_matrix.size()))
        self.transformation_matrix = transformation_matrix

    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)
        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__ + ‘(‘
        format_string += (str(self.transformation_matrix.numpy().tolist()) + ‘)‘)
        return format_string

ColorJitter類也比較常用，主要是修改輸入圖像的4大參數值：brightness, contrast and saturation，hue，也就是亮度，對比度，飽和度和色度。可以根據註釋來合理設置這4個參數。

class ColorJitter(object):
    """Randomly change the brightness, contrast and saturation of an image.

    Args:
        brightness (float): How much to jitter brightness. brightness_factor
            is chosen uniformly from [max(0, 1 - brightness), 1 + brightness].
        contrast (float): How much to jitter contrast. contrast_factor
            is chosen uniformly from [max(0, 1 - contrast), 1 + contrast].
        saturation (float): How much to jitter saturation. saturation_factor
            is chosen uniformly from [max(0, 1 - saturation), 1 + saturation].
        hue(float): How much to jitter hue. hue_factor is chosen uniformly from
            [-hue, hue]. Should be >=0 and <= 0.5.
    """
    def __init__(self, brightness=0, contrast=0, saturation=0, hue=0):
        self.brightness = brightness
        self.contrast = contrast
        self.saturation = saturation
        self.hue = hue

    @staticmethod
    def get_params(brightness, contrast, saturation, hue):
        """Get a randomized transform to be applied on image.

        Arguments are same as that of __init__.

        Returns:
            Transform which randomly adjusts brightness, contrast and
            saturation in a random order.
        """
        transforms = []
        if brightness > 0:
            brightness_factor = np.random.uniform(max(0, 1 - brightness), 1 + brightness)
            transforms.append(Lambda(lambda img: F.adjust_brightness(img, brightness_factor)))

        if contrast > 0:
            contrast_factor = np.random.uniform(max(0, 1 - contrast), 1 + contrast)
            transforms.append(Lambda(lambda img: F.adjust_contrast(img, contrast_factor)))

        if saturation > 0:
            saturation_factor = np.random.uniform(max(0, 1 - saturation), 1 + saturation)
            transforms.append(Lambda(lambda img: F.adjust_saturation(img, saturation_factor)))

        if hue > 0:
            hue_factor = np.random.uniform(-hue, hue)
            transforms.append(Lambda(lambda img: F.adjust_hue(img, hue_factor)))

        np.random.shuffle(transforms)
        transform = Compose(transforms)

        return transform

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Input image.

        Returns:
            PIL Image: Color jittered image.
        """
        transform = self.get_params(self.brightness, self.contrast,
                                    self.saturation, self.hue)
        return transform(img)

    def __repr__(self):
        return self.__class__.__name__ + ‘()‘

RandomRotation類是隨機旋轉輸入圖像，也比較常用，具體參數可以看註釋，在F.rotate()中主要是調用PIL Image的rotate方法。

class RandomRotation(object):
    """Rotate the image by angle.

    Args:
        degrees (sequence or float or int): Range of degrees to select from.
            If degrees is a number instead of sequence like (min, max), the range of degrees
            will be (-degrees, +degrees).
        resample ({PIL.Image.NEAREST, PIL.Image.BILINEAR, PIL.Image.BICUBIC}, optional):
            An optional resampling filter.
            See http://pillow.readthedocs.io/en/3.4.x/handbook/concepts.html#filters
            If omitted, or if the image has mode "1" or "P", it is set to PIL.Image.NEAREST.
        expand (bool, optional): Optional expansion flag.
            If true, expands the output to make it large enough to hold the entire rotated image.
            If false or omitted, make the output image the same size as the input image.
            Note that the expand flag assumes rotation around the center and no translation.
        center (2-tuple, optional): Optional center of rotation.
            Origin is the upper left corner.
            Default is the center of the image.
    """

    def __init__(self, degrees, resample=False, expand=False, center=None):
        if isinstance(degrees, numbers.Number):
            if degrees < 0:
                raise ValueError("If degrees is a single number, it must be positive.")
            self.degrees = (-degrees, degrees)
        else:
            if len(degrees) != 2:
                raise ValueError("If degrees is a sequence, it must be of len 2.")
            self.degrees = degrees

        self.resample = resample
        self.expand = expand
        self.center = center

    @staticmethod
    def get_params(degrees):
        """Get parameters for ``rotate`` for a random rotation.

        Returns:
            sequence: params to be passed to ``rotate`` for random rotation.
        """
        angle = np.random.uniform(degrees[0], degrees[1])

        return angle

    def __call__(self, img):
        """
            img (PIL Image): Image to be rotated.

        Returns:
            PIL Image: Rotated image.
        """

        angle = self.get_params(self.degrees)

        return F.rotate(img, angle, self.resample, self.expand, self.center)

    def __repr__(self):
        return self.__class__.__name__ + ‘(degrees={0})‘.format(self.degrees)

Grayscale類是用來將輸入圖像轉成灰度圖的，這裏根據參數num_output_channels的不同有兩種轉換方式。

class Grayscale(object):
    """Convert image to grayscale.

    Args:
        num_output_channels (int): (1 or 3) number of channels desired for output image

    Returns:
        PIL Image: Grayscale version of the input.
        - If num_output_channels == 1 : returned image is single channel
        - If num_output_channels == 3 : returned image is 3 channel with r == g == b

    """

    def __init__(self, num_output_channels=1):
        self.num_output_channels = num_output_channels

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be converted to grayscale.

        Returns:
            PIL Image: Randomly grayscaled image.
        """
        return F.to_grayscale(img, num_output_channels=self.num_output_channels)

    def __repr__(self):
        return self.__class__.__name__ + ‘()‘

RandomGrayscale類和前面的Grayscale類類似，只不過變成了按照指定的概率進行轉換。

class RandomGrayscale(object):
    """Randomly convert image to grayscale with a probability of p (default 0.1).

    Args:
        p (float): probability that image should be converted to grayscale.

    Returns:
        PIL Image: Grayscale version of the input image with probability p and unchanged
        with probability (1-p).
        - If input image is 1 channel: grayscale version is 1 channel
        - If input image is 3 channel: grayscale version is 3 channel with r == g == b

    """

    def __init__(self, p=0.1):
        self.p = p

    def __call__(self, img):
        """
        Args:
            img (PIL Image): Image to be converted to grayscale.

        Returns:
            PIL Image: Randomly grayscaled image.
        """
        num_output_channels = 1 if img.mode == ‘L‘ else 3
        if random.random() < self.p:
            return F.to_grayscale(img, num_output_channels=num_output_channels)
        return img

    def __repr__(self):
        return self.__class__.__name__ + ‘()‘

        </div>
                    <link href="https://csdnimg.cn/release/phoenix/mdeditor/markdown_views-7f770a53f2.css" rel="stylesheet">
            </div>

PyTorch源碼解讀之torchvision.transforms（轉）