/// <summary>
/// Constructs a CyclicNetwork with the provided neural net definition.
/// </summary>
public NeuralNetCyclic(
DirectedGraph digraph,
double[] weightArr,
VecFnSegment2 <double> activationFn,
int cyclesPerActivation)
{
Debug.Assert(digraph.ConnectionIdArrays._sourceIdArr.Length == weightArr.Length);
// Store refs to network structure data.
_srcIdArr = digraph.ConnectionIdArrays._sourceIdArr;
_tgtIdArr = digraph.ConnectionIdArrays._targetIdArr;
_weightArr = weightArr;
// Store network activation function and parameters.
_activationFn = activationFn;
_cyclesPerActivation = cyclesPerActivation;
// Store input/output node counts.
_inputCount = digraph.InputCount;
_outputCount = digraph.OutputCount;
// Create node pre- and post-activation signal arrays.
int nodeCount = digraph.TotalNodeCount;
_preActivationArr = new double[nodeCount];
_postActivationArr = new double[nodeCount];
// Wrap sub-ranges of the neuron signal arrays as input and output vectors.
this.InputVector = new VectorSegment <double>(_postActivationArr, 0, _inputCount);
// Note. Output neurons follow input neurons in the arrays.
this.OutputVector = new VectorSegment <double>(_postActivationArr, _inputCount, _outputCount);
}
private void RunBenchmark(IActivationFunction <double> actFn)
{
VecFnSegment2 <double> fn = actFn.Fn;
for (int i = 0; i < __loops; i++)
{
fn(_x, _w, 0, _x.Length);
}
}
/// <summary>
/// Constructs a CyclicNetwork with the provided neural net definition.
/// </summary>
public NeuralNetCyclic(
WeightedDirectedGraph <double> digraph,
VecFnSegment2 <double> activationFn,
int cyclesPerActivation)
: this(
digraph,
digraph.WeightArray,
activationFn,
cyclesPerActivation)
{
}
/// <summary>
/// Constructs a CyclicNetwork with the provided neural net definition.
/// </summary>
public CyclicNeuralNet(
WeightedDirectedGraph <double> digraph,
VecFnSegment2 <double> activationFn,
int activationCount,
bool boundedOutput)
: this(
digraph,
digraph.WeightArray,
activationFn,
activationCount,
boundedOutput)
{
}
/// <summary>
/// Constructs a CyclicNetwork with the provided neural net definition parameters.
/// </summary>
public CyclicNeuralNet(
WeightedDirectedGraph <double> digraph,
VecFnSegment2 <double> activationFn,
int activationCount,
bool boundedOutput)
{
// Store refs to network structure data.
_srcIdArr = digraph.ConnectionIdArrays._sourceIdArr;
_tgtIdArr = digraph.ConnectionIdArrays._targetIdArr;
_weightArr = digraph.WeightArray;
// Store network activation function and parameters.
_activationFn = activationFn;
_activationCount = activationCount;
// Store input/output node counts.
_inputCount = digraph.InputCount;
_outputCount = digraph.OutputCount;
// Create node pre- and post-activation signal arrays.
int nodeCount = digraph.TotalNodeCount;
_preActivationArr = new double[nodeCount];
_postActivationArr = new double[nodeCount];
// Wrap sub-ranges of the neuron signal arrays as input and output vectors.
_inputVector = new VectorSegment <double>(_postActivationArr, 0, _inputCount);
// Note. Output neurons follow input neurons in the arrays.
var outputVec = new VectorSegment <double>(_postActivationArr, _inputCount, _outputCount);
if (boundedOutput)
{
_outputVector = new BoundedVector(outputVec);
}
else
{
_outputVector = outputVec;
}
}
