Summary

This chapter focused on providing the reader with tools to handle different types of data and how to prepare them for the deep learning models. We started with time series data. This chapter next detailed how textual data needs to be preprocessed. This chapter showed how to perform data augmentation, an important technique for image classification and object detection. We next moved on to handling video; we show how to form image frames from a video. Next, this chapter covered audio files; we formed a time series and mel spectrogram from an audio file. Finally, we moved on to cloud platforms and discussed the features and services provided by three major cloud service providers.