MNIST MODEL Using Numpy and Maths ~ made by Sourasish Mondal

This is a basic Handwritten Number Recognition Model from the famous MNIST (Modified National Institute of Standards and Technology) dataset.

I have only used Python, Numpy and Mathematics to make this model. I used Pandas to read the data and matplotlib to show the predicted output

This Neural Network comprises of two simple layer architecture.

• Input Layer A^[0] will have 784 units corresponding to the 784 pixels in each 28x28 input image.
  
• Hidden Layer ^A[1] will have 10 units with ReLU activation.
  
• Output Layer ^A[2] will have 10 units corresponding to the ten digit classes with softmax activation.
  

The math behind it is given below: