Mastering Computer Vision with TensorFlow 2.x

Visualizing the layers of a neural network

In this chapter, we've learned about how an image is transformed into edges and then into feature maps and that by doing this, the neural network is able to predict the classes by combining many of the feature maps. In the first few layers, the neural network visualizes lines and corners, whereas in the last few layers, the neural network recognizes complex patterns such as feature maps. This can be broken down into the following categories.

• Building a custom image classifier model and visualizing its layers
• Training an existing advanced image classifier model and visualizing its layers

Let's take a look at these categories.