(the slow download speed of cifar10 in torchvision has been solved) how to download and use torchvision import

---

background

---

    train_set = torchvision.datasets.CIFAR10(root='./data', train=True,
                                             download=True, transform=transform)

Super slow , Too slow to report errors directly ：

urllib.error.URLError: urlopen error [Errno 11001] getaddrinfo failed

---

Solution

---

The official website is super fast http://www.cs.toronto.edu/~kriz/cifar.html
Download the first one ：

Unzip it , Then the directory will look like this

cifar-10-python/cifar-10-batches-py/*

among * Number is the following files .

We just need to copy the above files to data Under the table of contents ：

data/cifar-10-batches-py/*

And then use torchvision Handle , And set up download=False that will do .

    train_set = torchvision.datasets.CIFAR10(root='./data', train=True,
                                             download=False, transform=transform)

success .