Preparing the data to feed deep learning models

In the previous chapter, we explained that AutoKeras is a framework that specializes in deep learning that uses neural networks as a learning engine. We also learned how to create end-to-end classifier/regressor models for the MNIST dataset of handwritten digits as input data. This dataset had already been preprocessed to be used by the model, which means all the images had the same attributes (same size, color, and so on), but this is not always the case.

Once we know what a tensor is, we are ready to learn how to feed our neural networks. Most of the data preprocessing techniques are domain-specific, and we will explain them in the following chapters when we need to use them in specific examples. But first, we will present some fundamentals that are the basis for each specific technique.

Data preprocessing operations for neural network models

In this section, we will look at some of the operations we can use to transform the...