UnicodeDecodeError: ‘gbk‘ codec can‘t decode byte 0xae in position 60:

解决办法

 file = open(file_path, 'r',encoding='utf-8')

class MyDataset(Dataset):
    def __init__(self, txt, data_path=None, transform=None, target_transform=None, loader=default_loader):
        super(MyDataset, self).__init__() # 对继承父类的属性初始化
        # 在__init__()方法中得到图像的路径，然后将图像路径组成一个数组
        file_path = data_path + txt
        file = open(file_path, 'r',encoding='utf-8')
        imgs = []
        for line in file:
            line = line.split()
            # print(line[0].rstrip(','))  # img
            # print(line[1].rstrip('\n'))  # label
            imgs.append((line[0].rstrip(','), line[1].rstrip('\n')))
        self.imgs = imgs
        self.transform = transform
        self.target_transform = target_transform
        self.loader = loader
        self.data_path = data_path

    def __getitem__(self, index):
        # 按照索引读取每个元素的具体内容
        imgName, label = self.imgs[index]
        imgPath = self.data_path + imgName
        img = self.loader(imgPath)
        if self.transform is not None:
            img = self.transform(img)  # 数据标签转换为Tensor
            label = torch.from_numpy(np.array(int(label)))
        return img, label

    def __len__(self):
        # 数据集的图片数量
        return len(self.imgs)