Loading Data for ML Applications In this notebook, we will implement code to load data for ML applications. Following the approach used by PyTorch, we will implement a Dataset class and a DataLoader class. The Dataset class will be used to load the data and the DataLoader class will be used to iterate over the data in batches.
We will test it on a simple image dataset.
The Dataset Class First, we need some way of representing how each individual item will be pre-processed as the dataloader iterates over the data. We will do this by creating a Dataset class. Since this class represents multiple items in our dataset, we will need to define the following special methods: