Write a Python program that learns how to recognize handwritten digits using stochastic gradient descent and the MNIST training data. I. Implement the NN as described at the URL above. Use stochastic gradient descent to train the NN. Try different values of the hyper-parameters, learning rate (0.01,0.1, and 1), batch size (online learning of size 1, mini-batch of size 20, and full-batch of all instances), and epochs. Report the results achieved, classification accuracy, and training time (convergence speed), and make your conclusion regarding the hyperparameters effect on the learning process. II. Try creating a network with just two layers - an input and an output layer, no hidden layer - with 784 and 10 neurons, respectively. Train the network using stochastic gradient descent. What classification accuracy can you achieve