This is a simple Neural Network created without any Machine Learning Libraries. The only dependencies are the random module for a randomized weight and Matplotlib for visualization. The purpose of this Neural Network is to fit a line to a set of datapoints. I made this to better understand optimization and gradient descent.
- The Neural Network has a somewhat hard time with datasets larger than 20. (This can be fixed by lowering the learning rate)
- The Neural Network requires a lot of epochs to accurately calculate the y intercept