A basic neural network made in Python with numpy. It attempts to identify a colour
from an RGB value using the interact function.
The file colour_neural_net_v2.py provides the ColourNetwork class which,
after training (ie. using the train_from_file function) allows the user to
interact with the network. In doing so, you provide the ColourNetwork with an
RGB list (ie. [255, 0, 0] will be pure red) and it attempts to guess what the
colour is named.
Below is an example of usage:
>>> a = ColourNetwork()
>>> a.train_from_file("training1.txt")
>>> a.interact([33, 33, 200]) # A nice blue
Red: 0.0%
Orange: 0.0%
Yellow: 0.0%
Green: 0.0%
Blue: 100.0%
Indigo: 0.0%
Violet: 0.0%
Black: 0.0%
White: 0.0%
Best idea: Blue