public int Predict(List <float> input)
{
input.Add(1.0f);
float[] inputs = input.ToArray();
float[] values =
{
_n1.Activate(inputs),
_n2.Activate(inputs),
_n3.Activate(inputs)
};
float sum = 0;
foreach (var value in values)
{
sum += Mathf.Exp(value);
}
for (var i = 0; i < values.Length; i++)
{
values[i] = Mathf.Exp(values[i]) / sum;
}
var max = values.Max();
return(values.ToList().IndexOf(max));
}
public void Should_propogate_acitivation_energy()
{
var NeuronA = new Neuron();
var NeuronB = new Neuron();
NeuronA.Connect(NeuronB, 2.0);
NeuronA.Activate(2);
Assert.AreEqual(NeuronB.ActivatedValue, 4.0);
}
private double Activate(Pattern pattern)
{
for (int i = 0; i < pattern.Inputs.Length; i++)
{
_inputs[i].Output = pattern.Inputs[i];
}
foreach (Neuron neuron in _hidden)
{
neuron.Activate();
}
_output.Activate();
return(_output.Output);
}
public void Scaffold_Test_Neuron()
{
activator = new ReLU();
for (int i = 0; i < testInputs.Length; i++)
{
testInputs[i] = new Neuron();
}
neuronToTest = new Neuron(testInputs, activator, 2); // Defines each weight in the inputs as 1 for testing purposes
foreach (Neuron inputNeuron in testInputs)
{
inputNeuron.Input = 1;
inputNeuron.Activate(); // playing the role of the neuron layer
}
// Act for every test, test case
neuronToTest.Input = givenInput;
neuronToTest.Activate();
neuronToTest.SetError(expected);
}
public void InheritActivation()
{
neuronToTest.Input = null;
neuronToTest.Activate();
Assert.AreEqual(11, neuronToTest.Activation); // should give us the weighted sum!
}
public void Is_high_if_both_high()
{
_nA.Activate(1);
_nB.Activate(1);
Assert.AreEqual(_nOutput.ActivatedValue, 1);
}
