walkthrough.md

First, we have the nn directory which holds all neural network related classes like the modules, losses, optimizers and training utilities.

In the nn directory you will find 3 directories (the number of directories may change in the future) which are losses, optimizers and modules.

1. All of the essential neural network modules exists in the module directory like Conv1D, Conv2D, Linear, Dropout, etc..., if there is multiple modules that does a certain functionality it will be in a seperate directory inside the module directory (e.g. Convolution operations like Conv1D and Conv2D exists in the convolution directory.)

2. All the essential neural network optimizers exists in the optimizers directory. base_optimizer.py is the base optimizer class that all of our optimizers inheret from.

3. All the essential neural network losses exists in the losses directory. base_loss.py is the base loss class for all of our losses inheret from.