private static void SingleInput_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
net.InputVector[0] = 0.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(0.5, net.OutputVector[0]);
}
// Activate and test.
net.InputVector[0] = 1.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(actFn.Fn(1), net.OutputVector[0]);
}
// Activate and test.
net.InputVector[0] = 10.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(actFn.Fn(10), net.OutputVector[0]);
}
}
private static void TestOverloads(IActivationFunction <double> actFn, Random rng)
{
// Init an array of random values.
double[] v = GetRandomArray(1001, rng);
// Init empty target arrays.
double[] v_scalar = new double[v.Length];
double[] v_vector = new double[v.Length];
double[] v_zero = new double[v.Length];
// Apply the scalar overload (single input) of the activation function.
ApplyFunc(actFn.Fn, v, v_scalar);
// Apply vector based overloads.
// Overload 1.
Array.Copy(v, v_vector, v.Length);
actFn.Fn(v_vector);
Compare(v_scalar, v_vector);
// Overload 2.
Array.Copy(v, v_vector, v.Length);
actFn.Fn(v_vector, 10, 20);
Compare(v, v_vector, 0, 10);
Compare(v_scalar, v_vector, 10, 20);
Compare(v, v_vector, 20, v.Length);
// Overload 3.
Array.Clear(v_vector, 0, v_vector.Length);
actFn.Fn(v, v_vector, 10, 20);
Compare(v_zero, v_vector, 0, 10);
Compare(v_scalar, v_vector, 10, 20);
Compare(v_zero, v_vector, 20, v.Length);
}
private static void TwoInputs_WeightHalf_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
// Activate and test.
net.InputVector[0] = 0.0;
net.InputVector[1] = 0.0;
net.Activate();
Assert.Equal(0.5, net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 1.0;
net.InputVector[1] = 2.0;
net.Activate();
x = 1.5; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 10.0;
net.InputVector[1] = 20.0;
net.Activate();
x = 15.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
}
private static void TwoInputs_WeightHalf_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
// Activate and test.
inputs[0] = 0.0;
inputs[1] = 0.0;
net.Activate();
Assert.Equal(0.5, outputs[0]);
// Activate and test.
inputs[0] = 1.0;
inputs[1] = 2.0;
net.Activate();
x = 1.5; actFn.Fn(ref x);
Assert.Equal(x, outputs[0]);
// Activate and test.
inputs[0] = 10.0;
inputs[1] = 20.0;
net.Activate();
x = 15.0; actFn.Fn(ref x);
Assert.Equal(x, outputs[0]);
}
private static void MultipleInputsOutputs_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
net.InputVector[0] = 1.0;
net.InputVector[1] = 2.0;
net.InputVector[2] = 3.0;
net.Activate();
Assert.AreEqual(actFn.Fn(2.0), net.OutputVector[0]);
Assert.AreEqual(actFn.Fn(3.0), net.OutputVector[1]);
Assert.AreEqual(actFn.Fn(1.0), net.OutputVector[2]);
}
private static void ComplexCyclic_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Simulate network in C# and compare calculated outputs with actual network outputs.
double[] preArr = new double[3];
double[] postArr = new double[3];
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
postArr[0] = 3.0;
inputs[0] = 3.0;
for (int i = 0; i < 10; i++)
{
preArr[1] = postArr[0] * -2.0 + postArr[2];
preArr[2] = postArr[0] + postArr[1];
actFn.Fn(ref preArr[1], ref postArr[1]);
actFn.Fn(ref preArr[2], ref postArr[2]);
net.Activate();
Assert.Equal(postArr[1], outputs[0]);
}
// Rest the network's internal state.
net.Reset();
Assert.Equal(0.0, outputs[0]);
// Run the test again.
Array.Clear(preArr, 0, preArr.Length);
Array.Clear(postArr, 0, postArr.Length);
postArr[0] = 3.0;
inputs[0] = 3.0;
for (int i = 0; i < 10; i++)
{
preArr[1] = postArr[0] * -2.0 + postArr[2];
preArr[2] = postArr[0] + postArr[1];
actFn.Fn(ref preArr[1], ref postArr[1]);
actFn.Fn(ref preArr[2], ref postArr[2]);
net.Activate();
Assert.Equal(postArr[1], outputs[0]);
}
}
private static void Fn_Overloads_Inner(IActivationFunction <double> actFn, Random rng)
{
// Init an array of random values.
double[] v = GetRandomArray(1001, rng);
// Init empty target arrays.
double[] v_scalar = new double[v.Length];
double[] v_vector = new double[v.Length];
double[] v_zero = new double[v.Length];
// Apply the scalar overload (single input) of the activation function.
ApplyFunc(actFn.Fn, v, v_scalar);
// Apply vector based overloads.
// Overload 1.
Array.Copy(v, v_vector, v.Length);
actFn.Fn(v_vector);
Assert.Equal(v_scalar, v_vector);
// Overload 2.
Array.Copy(v, v_vector, v.Length);
actFn.Fn(v_vector.AsSpan(10..20));
ConponentwiseEqual(v, v_vector, 0, 10);
ConponentwiseEqual(v_scalar, v_vector, 10, 20);
ConponentwiseEqual(v, v_vector, 20, v.Length);
// Overload 3.
Array.Clear(v_vector, 0, v_vector.Length);
actFn.Fn(v[10..20], v_vector.AsSpan(10..20));
private static void HiddenNode_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
net.InputVector[0] = 0.0;
net.InputVector[1] = 0.0;
net.Activate();
Assert.AreEqual(actFn.Fn(1.0), net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 0.5;
net.InputVector[1] = 0.25;
net.Activate();
Assert.AreEqual(actFn.Fn(actFn.Fn(0.375) * 2.0), net.OutputVector[0]);
}
private static void Complex_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
net.InputVector[0] = 0.5;
net.InputVector[1] = 0.25;
net.Activate();
double output1 = actFn.Fn(actFn.Fn(0.25));
Assert.AreEqual(output1, net.OutputVector[1]);
double output0 = actFn.Fn(actFn.Fn(output1 + 0.5 + 0.25) * 0.9);
Assert.AreEqual(output0, net.OutputVector[0]);
}
private static void AssertMonotonic(IActivationFunction <double> actFn, bool strict)
{
Func <double, double> fn = delegate(double x) {
actFn.Fn(ref x);
return(x);
};
Assert.True(FuncTestUtils.IsMonotonicIncreasing(fn, -6, 6, 0.01, strict));
}
private static void MultipleInputsOutputs_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
// Activate and test.
net.InputVector[0] = 1.0;
net.InputVector[1] = 2.0;
net.InputVector[2] = 3.0;
net.Activate();
x = 2.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
x = 3.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[1]);
x = 1.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[2]);
}
private static void HiddenNode_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
// Activate and test.
net.InputVector[0] = 0.0;
net.InputVector[1] = 0.0;
net.Activate();
x = 1.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 0.5;
net.InputVector[1] = 0.25;
net.Activate();
x = 0.375; actFn.Fn(ref x);
x *= 2; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
}
private static void Complex_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
// Activate and test.
net.InputVector[0] = 0.5;
net.InputVector[1] = 0.25;
net.Activate();
x = 0.25; actFn.Fn(ref x); actFn.Fn(ref x);
double output1 = x;;
Assert.Equal(output1, net.OutputVector[1]);
x = output1 + 0.5 + 0.25; actFn.Fn(ref x);
x *= 0.9; actFn.Fn(ref x);
double output0 = x;
Assert.Equal(output0, net.OutputVector[0]);
}
private static void MultipleInputsOutputs_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
// Activate and test.
inputs[0] = 1.0;
inputs[1] = 2.0;
inputs[2] = 3.0;
net.Activate();
x = 2.0; actFn.Fn(ref x);
Assert.Equal(x, outputs[0]);
x = 3.0; actFn.Fn(ref x);
Assert.Equal(x, outputs[1]);
x = 1.0; actFn.Fn(ref x);
Assert.Equal(x, outputs[2]);
}
private static void HiddenNode_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
// Activate and test.
inputs[0] = 0.0;
inputs[1] = 0.0;
net.Activate();
x = 1.0; actFn.Fn(ref x);
Assert.Equal(x, outputs[0]);
// Activate and test.
inputs[0] = 0.5;
inputs[1] = 0.25;
net.Activate();
x = 0.375; actFn.Fn(ref x);
x *= 2; actFn.Fn(ref x);
Assert.Equal(x, outputs[0]);
}
private static void SingleInput_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
// Activate and test.
net.InputVector[0] = 0.0;
net.Activate();
Assert.Equal(0.5, net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 1.0;
net.Activate();
x = 1.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
// Activate and test.
net.InputVector[0] = 10.0;
net.Activate();
x = 10.0; actFn.Fn(ref x);
Assert.Equal(x, net.OutputVector[0]);
}
private static void SingleInput_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
// Activate and test.
inputs[0] = 0.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(0.5, outputs[0]);
}
// Activate and test.
double x;
x = 1.0; actFn.Fn(ref x);
inputs[0] = 1.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(x, outputs[0]);
}
// Activate and test.
x = 10.0; actFn.Fn(ref x);
inputs[0] = 10.0;
for (int i = 0; i < 10; i++)
{
net.Activate();
Assert.Equal(x, outputs[0]);
}
}
private static void Complex_WeightOne_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
double x;
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
// Activate and test.
inputs[0] = 0.5;
inputs[1] = 0.25;
net.Activate();
x = 0.25; actFn.Fn(ref x); actFn.Fn(ref x);
double output1 = x;
Assert.Equal(output1, outputs[1]);
x = output1 + 0.5 + 0.25; actFn.Fn(ref x);
x *= 0.9; actFn.Fn(ref x);
double output0 = x;
Assert.Equal(output0, outputs[0]);
}
private static void CyclicOutput_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
const double input = 0.1;
double inputVal = input;
net.InputVector[0] = inputVal;
for (int i = 0; i < 10; i++)
{
net.Activate();
double outputExpected = actFn.Fn(inputVal);
Assert.Equal(outputExpected, net.OutputVector[0]);
inputVal = input + outputExpected;
}
}
private static void CyclicOutput_Inner(
IBlackBox <double> net,
IActivationFunction <double> actFn)
{
// Activate and test.
const double input = 0.1;
var inputs = net.Inputs.Span;
var outputs = net.Outputs.Span;
double inputVal = input;
inputs[0] = inputVal;
for (int i = 0; i < 10; i++)
{
net.Activate();
double outputExpected = inputVal; actFn.Fn(ref outputExpected);
Assert.Equal(outputExpected, outputs[0]);
inputVal = input + outputExpected;
}
}