Deep Learning with PyTorch Quick Start Guide

Transforms

As well as the ToTensor() transform, the torchvision package includes a number of transforms specifically for Python imaging library images. We can apply multiple transforms to a dataset object using the compose function as follows:

Compose objects are essentially a list of transforms that can then be passed to the dataset as a single variable. It is important to note that the image transforms can only be applied to PIL image data, not tensors. Since transforms in a compose are applied in the order that they are listed, it is important that the ToTensor transform occurs last. If it is placed before the PIL transforms in the Compose ...