Description:
N2 is a neural network for the Palm. N2 requires MathLib.
N2 has the following characteristics:
- Transfer Function = Sigmoid range from -0.5 to 0.5
- Momentum = 0.9
- Number of Input Units = 2
- Number of Hidden Units = 2
- Number of Output Units = 1
N2 has four training patterns. Each pattern consists of two input units and one output unit.
The Train button performs forward and back propagation once for each training pattern and saves the weights.
The Run button uses the two test inputs to perform one forward propagation through the network.
The resulting output is displayed as a floating point value between -0.5 and 0.5.
