pytorch中的transforms介绍

transforms用法介绍

torchvision.transforms模块主要用于对图像进行转换等一系列预处理操作，其主要目的是对图像数据进行增强，进而提高模型的泛化能力。对图像预处理操作有数据中心化，缩放，裁剪，旋转，翻转，填充，添加噪声，灰度变换，线性变换，仿射变换，亮度，饱和度，对比变换等。

transforms.Compose

transforms.Compose是将一系列的图像转换函数进行组合，实现时能够按照这些函数的顺序依次去图像进行处理操作，需要注意的是同样的功能也可以用torch.nn.Sequential函数来实现。

CLASS torchvision.transforms.Compose(transforms)

transforms:表示图像变换组合的列

transforms.Compose具体实例的代码如下所示

transform_train = transforms.Compose([transforms.RandomCrop(cut_size),transforms.RandomHorizontalFlip(),transforms.ToTensor(),
])
transform_train = torch.nn.Sequential(transforms.RandomCrop(cut_size),transforms.RandomHorizontalFlip(),transforms.ToTensor(), )transform_test = transforms.Compose([transforms.TenCrop(cut_size),transforms.Lambda(lambda crops: torch.stack([transforms.ToTensor()(crop) for crop in crops])),
])
transform_test =  torch.nn.Sequential(transforms.TenCrop(cut_size),transforms.Lambda(lambda crops: torch.stack([transforms.ToTensor()(crop) for crop in crops])),)

transforms.ToTensor

transforms.ToTensor的作用是将一个PIL Image格式的图片或者是取值范围为[0,255][0,255][0,255]，形状为[H×W×C][\mathrm{H} \times \mathrm{W} \times \mathrm{C}][H×W×C]numpy.ndarray的数组转换为取值范围为[0.0,1.0][0.0,1.0][0.0,1.0]，形状为[C×H×W][\mathrm{C}\times \mathrm{H}\times \mathrm{W}][C×H×W]的tensor格式图片。

transforms.RandomCrop

transforms.RandomCrop的作用是在图片的随机位置上进行裁剪并返回新的图片。

CLASS torchvision.transforms.RandomCrop(size, padding=None, pad_if_needed=False, fill=0, padding_mode=‘constant’)

size：表示裁剪图片的输出尺寸，如果参数是一个整数则裁剪的是一个正方形

padding：表示图像每个边框上的可选填充。默认值是None

pad_if_needed：如果图像小于所需大小，它将填充图像，以避免引发异常

fill：表示像素填充值，默认值为0。如果元组长度为3，则用于分别填充R、G、B通道

padding_mode：表示像素填充值的类型，默认是常值，也有边缘填充，反射和对称

transforms.RandomCrop具体实例的代码实现和对应的可视化图如下所示

from pathlib import Path
import matplotlib.pyplot as plt
import numpy as np
import torch
import torchvision.transforms as T
from torchvision.io import read_imageplt.rcParams["savefig.bbox"] = 'tight'
torch.manual_seed(1)def show(imgs):fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)for i, img in enumerate(imgs):img = T.ToPILImage()(img.to('cpu'))axs[0, i].imshow(np.asarray(img))axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])plt.show()img1 = read_image(str(Path('assets') / 'KOBE1.png'))
img2 = read_image(str(Path('assets') / 'KOBE2.png'))
show([img1, img2])transforms = T.Compose([T.RandomCrop(224),])# transforms = torch.nn.Sequential(
#     T.RandomCrop(224),
# )device = 'cuda' if torch.cuda.is_available() else 'cpu'
img1 = img1.to(device)
img2 = img2.to(device)transformed_img1 = transforms(img1)
transformed_img2 = transforms(img2)
show([transformed_img1, transformed_img2])

transforms.RandomHorizontalFlip

transforms.RandomHorizontalFlip的作用是以特定的概率将图片进行水平翻转。

CLASS torchvision.transforms.RandomHorizontalFlip(p=0.5)

p：表示图片水平翻转的概率，默认值是0.5

transforms.RandomHorizontalFlip具体实例的代码实现和对应的可视化图如下所示

from pathlib import Path
import matplotlib.pyplot as plt
import numpy as np
import torch
import torchvision.transforms as T
from torchvision.io import read_imageplt.rcParams["savefig.bbox"] = 'tight'
torch.manual_seed(1)def show(imgs):fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)for i, img in enumerate(imgs):img = T.ToPILImage()(img.to('cpu'))axs[0, i].imshow(np.asarray(img))axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])plt.show()img1 = read_image(str(Path('assets') / 'KOBE1.png'))  # type : torch
img2 = read_image(str(Path('assets') / 'KOBE2.png'))  # type : torch
show([img1, img2])transforms = T.Compose([T.RandomHorizontalFlip(p=0.9),])# transforms = torch.nn.Sequential(
#      T.RandomHorizontalFlip(p=0.3),
# )device = 'cuda' if torch.cuda.is_available() else 'cpu'
img1 = img1.to(device)
img2 = img2.to(device)transformed_img1 = transforms(img1)
transformed_img2 = transforms(img2)
show([transformed_img1, transformed_img2])

transforms.TenCrop

transforms.RandomCrop的作用是可以将一张图片的四个角和中心进行裁剪后，然后加上返回的翻转后共10张图片，其中默认翻转是水平翻转。

CLASS torchvision.transforms.TenCrop(size, vertical_flip=False)

size：表示裁剪图片的输出尺寸，如果参数是一个整数则裁剪的是一个正方形

vertical_flip：表示图片是否用垂直翻转代替水平翻转None

需要注意的是transforms.TenCrop函数的输入必须是PIL\mathrm{PIL}PIL的图片格式，具体实例的代码实现和对应的可视化图如下所示

from PIL import Image
from pathlib import Path
import matplotlib.pyplot as plt
import numpy as npimport torch
import torchvision.transforms as Tplt.rcParams["savefig.bbox"] = 'tight'torch.manual_seed(0)def plot(imgs, with_orig=True, row_title=None, **imshow_kwargs):if not isinstance(imgs[0], list):# Make a 2d grid even if there's just 1 rowimgs = [imgs]num_rows = len(imgs)num_cols = len(imgs[0]) + with_origfig, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)for row_idx, row in enumerate(imgs):row = [orig_img] + row if with_orig else rowfor col_idx, img in enumerate(row):ax = axs[row_idx, col_idx]ax.imshow(np.asarray(img), **imshow_kwargs)ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])if with_orig:axs[0, 0].set(title='Original image')axs[0, 0].title.set_size(8)if row_title is not None:for row_idx in range(num_rows):axs[row_idx, 0].set(ylabel=row_title[row_idx])plt.tight_layout()plt.show()orig_img = Image.open(Path('assets') / 'KOBE1.png')   # tyep : PIL
(top_left, top_right, bottom_left, bottom_right, center, flip_top_left, flip_top_right, flip_bottom_left, flip_bottom_right, flip_center) = T.TenCrop(size=(200,200))(orig_img)
plot([[top_left, top_right, bottom_left, bottom_right, center], [flip_top_left, flip_top_right, flip_bottom_left, flip_bottom_right, flip_center]],with_orig=False)

格外需要注意transforms.TenCrop对于每张图片会返回10张变换后的图片，尤其是在测试阶段会导致图片数量和标签数量不匹配，可以进行如下处理

transform = Compose([FiveCrop(size), # this is a list of PIL ImagesLambda(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