Chapter 7. How to picture neural networks: in your head and on paper

 

In this chapter

  • Correlation summarization
  • Simplified visualization
  • Seeing the network predict
  • Visualizing using letters instead of pictures
  • Linking variables
  • The importance of visualization tools

“Numbers have an important story to tell. They rely on you to give them a clear and convincing voice.”

Stephen Few, IT innovator, teacher, and consultant

It’s time to simplify

It’s impractical to think about everything all the time. Mental tools can help

Chapter 6 finished with a code example that was quite impressive. Just the neural network contained 35 lines of incredibly dense code. Reading through it, it’s clear there’s a lot going on; and that code includes over 100 pages of concepts that, when combined, can predict whether it’s safe to cross the street.

I hope you’re continuing to rebuild these examples from memory in each chapter. As the examples get larger, this exercise becomes less about remembering specific letters of code and more about remembering concepts and then rebuilding the code based on those concepts.

Correlation summarization

The previously overcomplicated visualization

The simplified visualization

Simplifying even further

Let’s see this network predict

Visualizing using letters instead of pictures

Linking the variables

Everything side by side

The importance of visualization tools