Getting started

To get started simply clone/fork this repository or download the desired files. The files you'll need are neural_net.py and utils.py. NOTE: You need to have NumPy installed.

In the file you want to use the library simply add 'from neural_net import *' (which will in turn import utils) and you're good to go. For examples of usage take a look in the folder named examples which contains data sets, already trained networks and files with setup and testing.

You will probably be able to figure out how to use the library by looking at these examples and reading some docstrings provided that you have a basic understanding of neural networks, if you don’t, I recommend reading at least the first chapter in http://neuralnetworksanddeeplearning.com/.

Once you have trained a network you can use the methods save and load to avoid having to train the same network several times.

One important thing to note is the format of training/test data. This must be a python list of tuples with training examples. The first element of the tuple is the example input and the second is the corresponding output, both in the form of a vertical numpy array (ndarray with shape (n, 1)).

If you have any questions don't hesitate to open an issue. Good luck!