Preprocess images before feed into CNN

copyrightly · January 12, 2020, 4:54pm

I am doing some experiments on MNIST. I want to preprocess images before feed them into CNN. My preprocess function is like

def preprocess(image):
    ...
    return processed_image

the result is a processed image with the same size as the original image.
I am trying to modify this code: pytorch examples mnist
I am wondering where I should add this preprocess function into the code. Thank you!

Eta_C · January 13, 2020, 1:49am

github.com

pytorch/examples/blob/2796747f58516574f9fbdb2b62b709f034676060/mnist/main.py#L100-L112


train_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=True, download=True,
                   transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=args.batch_size, shuffle=True, **kwargs)
test_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=False, transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=args.test_batch_size, shuffle=True, **kwargs)

Write a transforms. Tutorial.