Sticker Breakout Detection
Here is the first Visual Basic program I've made that uses a NeuroShell 2's network: XOR
It's not quite finished and will never be. It's just an example of how to use neural networks to solve a problem.
What does it do ?
Well, it's really not complicated. The problem to solve is as follows: With given boolean inputs, what is the output of a XOR function ?
The XOR function table is:
Input 1 | Input 2 | Output
off | off | off
on | off | on
off | on | on
on | on | off
No neural network is required to calculate the output value, but it allows very easily to demonstrate the usage of a NN.
The problem could be something like: In the stairs there are two switches. One is upstairs, the other downstairs. If I push a given switch, should the light turn on ?
How to use it ?
- Choose a network: trained or not.
- Choose a representation: boolean or precise.
- Check the inputs to fire the network. The drawing represents the state of each neuron of the network.
- If the output is checked, the solution is "true", otherwise it's false.
If you choose the untrained network, the solution might be wrong. The "Expected" checkbox allows you to verify the rightness of the solution.
That's all for now.
Any comments, any questions, any ideas are welcomed. You can send me an email: [email protected] (well, not yet, soon hopefully, but you can come to see me anytime).