Preprocessing for VGG Network needed or not?

vision
1

Do we need to normalise the images to pass through vgg or they get preprocessed automatically?

If yes, please suggest a preprocessing script for images which is used generally?

2

To my understanding you need to resize your images and normalize the colors in them to get the best results. An example is as follows.

data_transform = transforms.Compose([
    transforms.Resize(224),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])]),

You would then pass this transform as a parameter for your datasets like this:

datasets.ImageFolder(path_to_training_directory, transform=data_transform)