"ConvNets are particularly well suited for data that has a spatial component to it. Things like images, audio, and text. Load the provided Fashion MNIST dataset from Tensorflow, split into training ...