L-layer-mnist-fashion-fully-numpy

L-layer mnist-fashion trained completely using numpy Neural network Without using any external library fashion mnist L-layer model. Mnist data contains 60,000 examples with 784 pixels each ranging from 1 to 255. Each training and test example is assigned to one of the following labels "T-shirt/top,Trouser, Pullover,Dress,Coat,Sandal,Shirt,Sneaker,Bag,Ankle boot" RELU and SGMOID activation functions are used,weights are initialized with he-et initialization. Adam Algorothm applied for model optimization.