An Artificial Neural Network created from scratch using Numpy and Pandas. This ANN has three layers.
The first layer(input) has 784 units corresponding to the 784 pixels in each digit input image.
The second layer(hidden) has 10 units with Rectified Linear Unit activation.
The third(output) layer has 10 units corresponding to the ten digit classes with softmax activation
The main motive behind creating this project is to implement and understand the underlying math behind a simple Artificial Neural Network.