The goal of this project will be to develop several ConvNets to correctly classify images to develop a scalable approach to image recognition. The approaches described can be applied to labeling the content of images, performing image searches, guiding robots, self-driving cars, and more.