In this book, we covered how to apply various deep learning networks to develop prediction and classification models. Several tips and tricks that we covered were unique to certain application areas and helped us arrive at better prediction or classification performance for the models that we developed.

In this chapter, we will go over certain tips and tricks that will be very handy when you continue your journey of applying these methods to new data and different problems. We will cover four topics in total. Note that these approaches haven't been covered in the previous chapters, but we will make use of some of the examples from them to illustrate their use.

In this chapter, we will cover the following topics:

• TensorBoard for training performance visualization
• Visualizing deep network models with LIME
• Visualizing model training with tfruns
• Early stopping of network training