chapter eight

8 Getting started with deep learning with tabular data

 

This chapter covers

  • Introduction to the deep learning with tabular data stacks - low-level frameworks and high-level application programming interfaces (APIs) for deep learning
  • The PyTorch with fastai stack
  • The PyTorch with TabNet stack
  • The PyTorch with Lightning Flash stack
  • Description of the stacks we didn’t exercise and why we didn’t exercise them
  • Comparison of the pros and cons of the deep learning with tabular data stacks

In the preceding four chapters (chapters 4 to 7), we have focused on machine learning with tabular data, that is, non-deep learning approaches to dealing with tabular data. This set of chapters culminated in chapter 7 where we went through an end-to-end example of applying a gradient boosting approach to the Kuala Lumpur real estate tabular dataset. For the next three chapters we will focus on deep learning approaches to tabular data, starting with the examination of deep learning with tabular data stacks in this chapter.

8.1 The deep learning with tabular data stack

8.2 PyTorch with fastai

8.2.1 Reviewing the key code aspects of the fastai solution

8.2.2 Comparing the fastai solution with the Keras solution

8.3 PyTorch with TabNet

8.3.1 Reviewing the key code aspects of the TabNet solution

8.3.2 Comparing the TabNet solution with the Keras solution

8.4 PyTorch with Lightning Flash

8.4.1 Reviewing the key code aspects of the Lightning Flash solution

8.4.2 Comparing the Lightning Flash solution with the Keras solution

8.5 Overall comparison of the stacks

8.6 The stacks we didn’t explore

8.7 Summary