One issue that I noticed is that my current code adds weights based on
neuron not based on the connections between them. For these very
simple examples it's cool, but it would need to be fixed if you want a
neuron to support connections of differing weights.
On Jan 29, 12:50 am, Justin Bozonier <darkxant...@gmail.com> wrote:
> I implemented a simple neural net class in C# that can be used to
> mimick the AND operator. I posted the code on GitHub here:
>