Deep-learning

The attached file implemented the Softmax Classifier and 100-hidden layer Neural Network for the random generating data, the function included backpropogation, loss function, and gradient descent. After implementation, the Softman Classifier has an accuracy of 52 percent and the Neural Network has an accuracy of 99 percent. I have also looked through assignment two but it seems like I need to do a lot of setup in google drive colloborator to access the functions and implement them. As I am not particuraly good at set ups and time is limited, I checked through some answers in github and get a taste of the how the filters work in CNN. That's what I have done for the coding part.