Data preparation

- [Instructor] Data preparation is the first step in developing a deep learning model. This step consists of loading the data, applying transforms, and batching the data using PyTorch built-in capabilities. We won't carry how to generate a good dataset, as we use an existing popular academic dataset called CIFAR-10, developed by researchers from the Canadian Institute for Advanced Research, or short, CIFAR. CIFAR-10 dataset is a subset of a much larger dataset with 80 million images in it. It consists of 60,000 small color photographs of objects from 10 classes divided into 50,000 training images and 10,000 test images. Here is the table with class labels and their associated integer values. We use a Python library called Torchvision, as it has classes that support computer vision. The Torchvision datasets module provides several subclasses to load image data from standard data sets, such as our CIFAR-10 dataset. To create a training dataset using the existing CIFAR-10 dataset we'll…