Basic Neural Network from Scratch

I finished my implementation of backpropagation. I added batches and averaged. My code is really messy and not adaptable (it’s incredibly difficult to change the number of layers in the network) so I will soon create an array with the weights and biases in each layer so I can loop over them easily.

Implemented backpropagation. It’s highly unoptimized and poorly written because I barely understand the whole thing and I suck at C++. Tomorrow I hope to optimize my code.

First, I implemented a cost function for the network based on its identification. Next, I realized that my code would be much cleaner if I created a network class, so I restructured my code. In the class, I implemented a randomize function and edited my identify function. Since I barely knew any C++ coming into this project, I quickly realized that my OOP skills were lacking so I watched a few tutorials on it. I also revised backpropagation and will hopefully implement it tomorrow.

Reorganized some files and added weights and biases. Implemented random libraries and started organized my files for training.