People,
I have an idea which I hope will help me understand Neural Networks better - and I wonder if anyone who understands NNs would be interested in helping me to get a shared Google spreadsheet of a very simple NN going. For example the SS could have:
- two input nodes
- three hidden nodes
- one output node
all represented on a block of lines where I can copy this block to another block directly below the first block - representing each iteration of a loop in a program - then I can see easily by direct comparison what happens to the set of numbers with each iteration (ie each copied block).
What do people think? Is this doable? I looked around and found some Excel example NNs but I don't think they will import into Google Calc - and it would be better for my understanding to do it from scratch anyway . .
If we could get it going, it would be Open Source of course and it might be useful for other people who want to understand NNs by looking at a simple example in a familiar environment?
Thanks,
Phil.