I'm really surprised at how fast the first attempt went. I thought I would struggle a bit more because of all the details I don't know. Sure, I know that there are layers and some are hidden. I know that you can have different kinds of layers Sigmoid and Tanh, but I don't know the difference yet. In spite of what I don't know, I was still able to follow the very simplistic tutorial.
Step One: Start Python
Step One: Start Python
Step Two: Import pybrain and create a network.
With a network that has 3 hidden layers, 2 inputs, and 1 output neuron, would you believe that the next step is the neural network equivalent of a "Hello World" program? The modules are already connected and ready to do something.
Step Three: Activate the Network
Step Three: Activate the Network
Done. What do those numbers mean? Nothing. We didn't use any training data and nothing is truly ready, but just as in a "Hello World" tutorial, this is just to show that something has been accomplished. There is so much more to the subject, but until I read some books on deep learning, I think I've reached the end of what I can do without knowledge. It's a fun first step.
