In this chapter, you learned the basics of convolution and why it is an effective mechanism for image label prediction. You learned about basic concepts such as strides
and padding. This was followed by an example based on the Stanford dataset of Cats versus Dogs. We used three convolution layers to build the neural network and two fully connected layers to showcase how it is used to classify the images. We also plotted the weights for three layers and saw how filters modify the image. We also looked at concepts such as image pooling and how it helps make CNN more efficient.
In the next chapter we look at a different kind of neural network called a Recurrent Neural Network (RNN), which processes time series data or is used for natural language processing (NLP) to predict next word in a sequence