这段代码的类取自xiaodong这里：https://github.com/xiaodongww/pytorch/blob/master/cifarDataset.py
类：这是原始https://github.com/pytorch/vision/blob/master/torchvision/datasets/cifar.py的修改版本，为cifar100添加了粗糙标签作为https://github.com/xiaodongww/pytorch/blob/master/cifarDataset.py选项

from __future__ import print_function
from PIL import Image
import os
import os.path
import errno
import numpy as np
import sys
if sys.version_info[0] == 2:
    import cPickle as pickle
else:
    import pickle

import torch.utils.data as data
from torchvision.datasets.utils import download_url, check_integrity

class CIFAR10(data.Dataset):
    """`CIFAR10 <https://www.cs.toronto.edu/~kriz/cifar.html>`_ Dataset.
    Args:
        root (string): Root directory of dataset where directory
            ``cifar-10-batches-py`` exists.
        train (bool, optional): If True, creates dataset from training set, otherwise
            creates from test set.
        transform (callable, optional): A function/transform that  takes in an PIL image
            and returns a transformed version. E.g, ``transforms.RandomCrop``
        target_transform (callable, optional): A function/transform that takes in the
            target and transforms it.
        download (bool, optional): If true, downloads the dataset from the internet and
            puts it in root directory. If dataset is already downloaded, it is not
            downloaded again.
    """
    base_folder = 'cifar-10-batches-py'
    url = "https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz"
    filename = "cifar-10-python.tar.gz"
    tgz_md5 = 'c58f30108f718f92721af3b95e74349a'
    train_list = [
        ['data_batch_1', 'c99cafc152244af753f735de768cd75f'],
        ['data_batch_2', 'd4bba439e000b95fd0a9bffe97cbabec'],
        ['data_batch_3', '54ebc095f3ab1f0389bbae665268c751'],
        ['data_batch_4', '634d18415352ddfa80567beed471001a'],
        ['data_batch_5', '482c414d41f54cd18b22e5b47cb7c3cb'],
    ]

    test_list = [
        ['test_batch', '40351d587109b95175f43aff81a1287e'],
    ]

    def __init__(self, root, train=True,
                 transform=None, target_transform=None,
                 download=False, coarse=False):
        self.root = os.path.expanduser(root)
        self.transform = transform
        self.target_transform = target_transform
        self.train = train  # training set or test set
        self.coarse = coarse

        if download:
            self.download()

        if not self._check_integrity():
            raise RuntimeError('Dataset not found or corrupted.' +
                               ' You can use download=True to download it')

        # now load the picked numpy arrays
        if self.train:
            self.train_data = []
            self.train_labels = []
            self.train_coarse_labels = []
            for fentry in self.train_list:
                f = fentry[0]
                file = os.path.join(self.root, self.base_folder, f)
                fo = open(file, 'rb')
                if sys.version_info[0] == 2:
                    entry = pickle.load(fo)
                else:
                    entry = pickle.load(fo, encoding='latin1')
                self.train_data.append(entry['data'])
                if 'labels' in entry:
                    self.train_labels += entry['labels']
                else:
                    self.train_labels += entry['fine_labels']
                    if self.coarse:
                        self.train_coarse_labels += entry['coarse_labels']
                fo.close()

            self.train_data = np.concatenate(self.train_data)
            self.train_data = self.train_data.reshape((50000, 3, 32, 32))
            self.train_data = self.train_data.transpose((0, 2, 3, 1))  # convert to HWC
        else:
            f = self.test_list[0][0]
            file = os.path.join(self.root, self.base_folder, f)
            fo = open(file, 'rb')
            if sys.version_info[0] == 2:
                entry = pickle.load(fo)
            else:
                entry = pickle.load(fo, encoding='latin1')
            self.test_data = entry['data']
            if 'labels' in entry:
                self.test_labels = entry['labels']
            else:
                self.test_labels = entry['fine_labels']
                if self.coarse:
                    self.test_coarse_labels = entry['coarse_labels']
            fo.close()
            self.test_data = self.test_data.reshape((10000, 3, 32, 32))
            self.test_data = self.test_data.transpose((0, 2, 3, 1))  # convert to HWC

    def __getitem__(self, index):
        """
        Args:
            index (int): Index
        Returns:
            tuple: (image, target) where target is index of the target class.
        """
        if self.train:
            img, target = self.train_data[index], self.train_labels[index]
            if self.coarse:
                coarse_target = self.train_coarse_labels[index]
        else:
            img, target = self.test_data[index], self.test_labels[index]
            if self.coarse:
                coarse_target = self.test_coarse_labels[index]

        # doing this so that it is consistent with all other datasets
        # to return a PIL Image
        img = Image.fromarray(img)

        if self.transform is not None:
            img = self.transform(img)

        if self.target_transform is not None:
            target = self.target_transform(target)
        if not self.coarse:
            return img, target
        else:
            return img, target, coarse_target

    def __len__(self):
        if self.train:
            return len(self.train_data)
        else:
            return len(self.test_data)

    def _check_integrity(self):
        root = self.root
        for fentry in (self.train_list + self.test_list):
            filename, md5 = fentry[0], fentry[1]
            fpath = os.path.join(root, self.base_folder, filename)
            if not check_integrity(fpath, md5):
                return False
        return True

    def download(self):
        import tarfile

        if self._check_integrity():
            print('Files already downloaded and verified')
            return

        root = self.root
        download_url(self.url, root, self.filename, self.tgz_md5)

        # extract file
        cwd = os.getcwd()
        tar = tarfile.open(os.path.join(root, self.filename), "r:gz")
        os.chdir(root)
        tar.extractall()
        tar.close()
        os.chdir(cwd)

class CIFAR100(CIFAR10):
    """`CIFAR100 <https://www.cs.toronto.edu/~kriz/cifar.html>`_ Dataset.
    This is a subclass of the `CIFAR10` Dataset.
    """
    base_folder = 'cifar-100-python'
    url = "https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz"
    filename = "cifar-100-python.tar.gz"
    tgz_md5 = 'eb9058c3a382ffc7106e4002c42a8d85'
    train_list = [
        ['train', '16019d7e3df5f24257cddd939b257f8d'],
    ]

    test_list = [
        ['test', 'f0ef6b0ae62326f3e7ffdfab6717acfc'],
    ]

放入train_data和test_data变量：

train_data = CIFAR100('./data', train=True,
                 transform=transform_train,
                 download=True, coarse=True)
test_data = CIFAR100('./data', train=False,
                 transform=transform_test,
                 download=True, coarse=True)

Data loader:

    train_length = train_data.__len__() # Length training dataset
    train_indices = np.arange(train_length) # Create arange 
    np.random.shuffle(train_indices) # Randomly Suffle training indices
    
    
    
    train_loader = DataLoader(
                            train_data, 
                            batch_size=batch_size, 
                            sampler = SubsetRandomSampler(train_indices[:int(train_length*0.8)]) # 80% of data in train set
                        )
    
    val_loader = DataLoader(
                            train_data, 
                            batch_size=batch_size, 
                            sampler = SubsetRandomSampler(train_indices[int(train_length*0.8):]) #20% of data in validation set
                        )
    
    test_loader = DataLoader(
                            test_data, 
                            batch_size=batch_size,
                        )
    
    
    # Disable Data Augmentation on Validation Set
    val_loader.dataset.transform = transform_test

您可以测试它是否与以下课程标签一起工作：

#Basic EDA with labels for train_data
demo_loader = torch.utils.data.DataLoader(train_data, batch_size=10)

batch = next(iter(demo_loader))
img, course_labels, fine_labels = batch
print(type(img), type(course_labels), type(fine_labels))
print(img.shape, course_labels.shape, fine_labels.shape)

您可以在此处打印课程标签和精细标签
我在培训中使用

for batch_idx, (data,targets, fine_targets) in enumerate(val_loader):
            data = data.to(DEVICE)
            targets = targets.to(DEVICE)
            fine_targets = fine_targets.to(DEVICE)

然后，您可以在训练循环或验证循环中使用fine_targets来访问它