In this chapter, you learned how to deal with having datasets that are too big to be handled by your normal desktop computer. We saw how to train TensorFlow models across multiple GPUs and machines, and finally, we looked at best practices for storing your data and feeding it to your model efficiently.
Over the course of this book, we have looked at many of the current popular problems in computer vision and how deep learning can be used to tackle all of them. We have also provided insight into how these might be implemented in TensorFlow. Along the way, we gave an introduction to how to use TensorFlow.