读取图片数据集：：
# -*- coding: utf-8 -*-
import torch as t
from torch.utils import data
import os
from PIL import Image
import numpy as np

class DogCat(data.Dataset):
 def __init__(self,root):
 imgs=os.listdir(root)
 #所有图片的绝对路径
 #这里不实际加载图片，只是指定路径，当调用__getitem__时才会真正读图片
 self.imgs=[os.path.join(root, img) for img in imgs]

 def __getitem__(self, index):
 img_path=self.imgs[index]
 #dog->1, cat->0
 label=1 if 'dog' in img_path.split("/")[-1] else 0
 pil_img=Image.open(img_path)
 array=np.asarray(pil_img)
 data=t.from_numpy(array)
 return data,label

 def __len__(self):
 return len(self.image)

dataset=DogCat('data/train')
img,label=dataset[0]#相当于调用dataset.__getitem__(0)
for img,label in dataset:
 print(img.size(),img.float().mean(),label)



第二：改变图片尺寸

#-*- coding: utf-8 -*-
import os
from PIL import Image
from torch.utils import data
import numpy as np
from torchvision import transforms as T


transforms=T.Compose([
 T.Resize(224), #缩放图片（Image）,保持长宽比不变，最短边为224像素
 T.CenterCrop(224), #从图片中间裁剪出224*224的图片
 T.ToTensor(), #将图片Image转换成Tensor，归一化至【0,1】
 T.Normalize(mean=[.5,.5,.5],std=[.5,.5,.5]) #标准化至【-1,1】，规定均值和方差
])

class DogCat(data.Dataset):
 def __init__(self,root, transforms=None):
 imgs=os.listdir(root)
 self.imgs=[os.path.join(root, img) for img in imgs]
 self.transforms=transforms

 def __getitem__(self, index):
 img_path=self.imgs[index]
 #dog->1, cat->0
 label=1 if 'dog' in img_path.split("/")[-1] else 0
 data=Image.open(img_path)
 if self.transforms:
 data=self.transforms(data)
 return data,label

 def __len__(self):
 return len(self.imgs)
dataset=DogCat('data/train', transforms=transforms)
img,label=dataset[0]#相当于调用dataset.__getitem__(0)
for img,label in dataset:
 print(img.size(),label)

#使用ImageFolder读取图片

#-*- coding: utf-8 -*-
from torchvision.datasets import ImageFolder
dataset=ImageFolder('data/')
print(dataset.class_to_idx)
print(dataset.imgs)