深度学习pytorch代码: transforms结构及用法 常见的transforms

transforms.py 工具箱

1. totensor 
2. resize

图片-------->工具(transforms)---------->结果

                       |

    使用工具创建具体的工具

     transforms.ToTensor()

使用工具：

        输入：图片

        输出：result = tool(图片)

from PIL import Image as imim
from tensorboardX import SummaryWriter
from torchvision import transforms
# python的用法 -》 tensor的数据类型
# 通过transform.ToTensor解决两个问题
# 1.transforms该如何让使用(python)
# 2.为什么需要ToTensor类型


img_path = r"C:\Users\123\Desktop\python4.7\hymenoptera_data\train\ants\6240338_93729615ec.jpg"
img = imim.open(img_path)
# print(img)


writer = SummaryWriter ("logs")
tensor_trans = transforms.ToTensor()
tensor_img = tensor_trans(img)    # ctrl+p  查看需要填入的类型
# print(tensor_img)
writer.add_image("Tensor_img", tensor_img)
writer.close()

常见的Transforms

* PIL           * Image.open()

* tensor      * ToTensor()

* narray      * cv.imread()

注意事项：

1. 关注输入和输出类型
2. 多看官方文档
3. 关注方法需要什么参数
4. 不知道返回值的时候  print / print(type()) / debug

from PIL import Image
from tensorboardX import SummaryWriter
from torchvision import transforms

writer = SummaryWriter("logs")
img = Image.open(r"C:\Users\123\Desktop\python4.7\flower.jpg")
print(img)


# ToTensor的使用
trans_totensor = transforms.ToTensor()  # 创建对象
img_tensor = trans_totensor(img)        # 调用方法
writer.add_image("ToTensor", img_tensor)


# Normalize
# output[channel] = (input[channel] - mean[channel]) / std[channel]  均值和方差
print(img_tensor[0][0][0])   # 像素的第一层第一行第一列
trans_norm = transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
img_norm = trans_norm(img_tensor)
print(img_norm[0][0][0])
writer.add_image("Normalize", img_norm)   # 绘制图片

# resize - 1
print(img.size)
trans_resize = transforms.Resize(512, 512)
# img PIL -> resize -> img_resize PIL
img_resize = trans_resize(img)
# img_resize PIL -> totensor -> img_resize tensor
image_resize = trans_totensor(img_resize)
print(img_resize)
writer.add_image("Resize-1", img_resize, 0)
print(img_resize)

# Compose - resize - 2
trans_resize_2 = transforms.Resize(512)
# PIL -> PIL -> tensor
trans_compose = transforms.Compose(
    [trans_resize_2, trans_totensor]
)
img_resize_2 = trans_compose(img)
writer.add_image("Resize-2", img_resize_2, 1)

# RandomCrop
trans_random = transforms.RandomCrop(512)
trans_compose_2 = transforms.Compose([trans_random, trans_totensor])
for i in range(10):   # 裁剪10个
    img_crop = trans_compose_2(img)
    writer.add_image("RandomCrop", img_crop, i)

writer.close()