The Convolutional Layer

A convolutional layer allows a neural net to learn more and more complex features. A picture can be either a 2d or 3d object depending on how you look at it. imagine that your picture is a bunch of bricks and They're assembled into a wall. You're sending the pixels into your neural net and then there's what's called a filter or a kernel And that's your piece of poster board So usually it's something like three pixels by three pixels or five pixels by five pixels, right?