CNNs, also known as ConvNets, are a class or a category of neural networks that are generally accepted to be very good at image classification, that is to say, they are very good at distinguishing cats from dogs, cars from planes, and many other common classification tasks.

A CNN typically consists of convolution layers, activation layers, and pooling layers. However, it has been structured specifically to take advantage of the fact that the inputs are typically images, and take advantage of the fact that some parts of the image are very likely to be next to each other.

They are actually fairly similar implementation wise to the feedforward networks that we have covered in earlier chapters.