Coding MNIST Neural Network from Scratch

Designed a neural network from scratch to classify handwritten digits from the MNIST dataset.

- The network is a 2 layer neural network with 128 hidden units and a sigmoid activation function.
- The network is trained using backpropagation and stochastic gradient descent.
- The network is evaluated on a test set of 10,000 images.
- The network achieves an accuracy of 97.4% on the test set.