chapter five
5 A Deeper Look at Deep Learning and the Deep Learning Stack
This chapter covers
- An overview of the end-to-end solution for the streetcar delay problem
- A deeper look at what distinguishes deep learning from non-deep learning machine learning
- What changed to make deep learning “work”?
- Using Pandas to accomplish the kind of operations you would typically do with SQL
- Introduction to Keras
- Comparison of Keras and TensorFlow 2.0
- Using embeddings to get the power of linear algebra for non-numeric values
- How the input dataset structure defines the Keras model
In the first four chapters of this book we reviewed the basic concepts of deep learning and structured data, introduced the Pandas library, and dug into the details of the primary example (streetcar delays) and how to prepare the sample dataset to train a deep learning model. In the next chapter we will go through the code and steps to prepare the model for training. To prepare for the Chapter 6 dive into the details of the deep learning model, in this chapter we’ll dig into concepts from deep learning and the deep learning stack (including Pandas and Keras).