public double FindMin(IFunction fun)
{
List <double[]> nests = new List <double[]>();
for (int i = 0; i < nestCount; i++)
{
nests.Add(RandGenerator.RandomArray(dimension, Space[0], Space[1]));
}
for (int i = 0; i < iteration; i++)
{
int NestIndex = rand.Next(0, nests.Count);
double[] flightResult = Move(nests[NestIndex], iterationWalk);
int cNestIndex = ChooseNest(NestIndex, nests);
if (fun.Calculate(flightResult) <= fun.Calculate(nests[cNestIndex]))
{
nests[cNestIndex] = flightResult;
}
nests = nests.OrderBy(x => fun.Calculate(x)).ToList();
Kill(nests, survivaleRate);
for (int j = nests.Count; j < nestCount; j++)
{
nests.Add(RandGenerator.RandomArray(dimension, Space[0], Space[1]));
}
}
nests = nests.OrderBy(x => fun.Calculate(x)).ToList();
return(fun.Calculate(nests[0]));
}
public double FindMin(IFunction function)
{
double[] currentState = RandGenerator.RandomArray
(dimensions, searchSpace[0], searchSpace[1]);
double T = maxTemp;
double alpha = 0.999;
for (int i = 0; i < iterTreshold; i++)
{
double[] newState = new double[dimensions];
currentState.CopyTo(newState, 0);
for (int j = 0; j < dimensions; j++)
{
newState[j] += RandGenerator.RandomNormal(-1.0, 1.0, 1) * T;
}
if (AcceptanceProb(function.Calculate(currentState), function.Calculate(newState), T) >= rand.NextDouble())
{
currentState = newState;
}
T *= alpha;
}
return(function.Calculate(currentState));
}
public double FindMin(IFunction fun)
{
List <Lift> population = new List <Lift>();
for (int i = 0; i < populations; i++)
{
Lift l = new Lift(RandGenerator.RandomArray(dimension, Space[0], Space[1]));
population.Add(l);
}
double[] bestPos = RandGenerator.RandomArray(dimension, Space[0], Space[1]);
double best = fun.Calculate(bestPos);
for (int l = 0; l < generations; l++)
{
for (int i = 0; i < populations; i++)
{
for (int j = 0; j < populations; j++)
{
if (i == j)
{
continue;
}
if (fun.Calculate(population[j].Pos) < fun.Calculate(population[i].Pos))
{
double dist = Calculate(population[i], population[j]);
double[] Position = new double[dimension];
population[i].Pos.CopyTo(Position, 0);
for (int k = 0; k < dimension; k++)
{
Position[k] += (population[j].Pos[k] - population[i].Pos[k]) * dist;
Position[k] += ((random.NextDouble() - 0.5) * rndFactor);
}
if (fun.Calculate(Position) < fun.Calculate(population[i].Pos))
{
population[i].Pos = Position;
}
}
}
double[] pos = population.Aggregate((x, y) => fun.Calculate(x.Pos) < fun.Calculate(y.Pos) ? x : y).Pos;
if (fun.Calculate(pos) < best)
{
bestPos = pos;
best = fun.Calculate(bestPos);
}
}
}
return(fun.Calculate(bestPos));
}
//protected void testResult(int nFunctionDimension, int mFunctionDimension, MeasuredPoint[] measuredPoints, Func<double[], double[]> func)
public void testResult(int nFunctionDimension, int mFunctionDimension, Func <double[], double[]> func)
{
IFunction model = (IFunction)approx;
Console.WriteLine("\n\nTest result approximation");
for (int i = 1; i < 10; i += 2)
{
double[] point = new double[nFunctionDimension + mFunctionDimension];
double[] pointX = new double[nFunctionDimension];
double[] pointY = new double[mFunctionDimension];
for (int j = 0; j < nFunctionDimension; j++)
{
point[j] = model.Min[j] + ((model.Max[j] - model.Min[j]) * i / 10);
pointX[j] = point[j];
}
model.Calculate(point);
for (int k = 0; k < mFunctionDimension; ++k)
{
pointY[k] = point[nFunctionDimension + k];
}
double[] realPointY = func(point);
double absErr = realPointY.Zip(pointY, (d1, d2) => Math.Abs(d1 - d2)).ToArray().Sum() / realPointY.Length;
Console.WriteLine("Vector " + String.Join(", ", pointX) + " Approximation result: " + String.Join(", ", pointY) +
", real: " + String.Join(", ", realPointY) + ", err: " + absErr);
}
}
public override void Execute(IFunction function, ResultRecord result) {
double value;
var votes = GetVotes().ToList();
function.Calculate(votes, result.ContentItemRecord.Id, out value);
result.Value = value;
result.Count = votes.Count;
}
/// <summary>
/// Peform the function execution
/// </summary>
/// <typeparam name="T"></typeparam>
/// <param name="function"></param>
/// <returns></returns>
private static object Execute <T>(IFunction <T> function)
{
function.ResolveInputs();
var rawOutput = function.Calculate();
return(function.Render(rawOutput));
}
/// <summary>
/// GetColor
/// </summary>
public Color GetColor(List <double> values, double alpha)
{
Color color;
var results = function.Calculate(values);
if (results != null)
{
color = alternateColorSpace.GetColor(results, alpha);
}
else
{
byte gray = 0xB0;
color = Color.FromArgb((byte)(alpha * 255), gray, gray, gray);
}
return(color);
}
public override void Execute(IFunction function, ResultRecord result)
{
double value;
var votes = GetVotes().ToList();
function.Calculate(votes, result.ContentItemRecord.Id, out value);
result.Value = value;
result.Count = votes.Count;
}
void PrintFunction(double start, double end, double step, IFunction function)
{
for (double x = start; x <= end; x += step)
{
try
{
PrintResult(x, function.Calculate(x), function);
}
catch (Exception err)
{
PrintErrorMessage(x, err.Message, function);
}
}
}
public double Calculate(double x, double t)
{
double omega = _parameterProvider.GetValue("omega");
double beta = _parameterProvider.GetValue("beta");
double res = 0;
for (int i = 0; i <= N; ++i)
{
res += _eFund.Calculate(i, x) * Math.Sqrt(omega)
* _laguerreFunc.Calculate(i, omega * t) * Math.Exp(-(beta / 2.0) * t);
}
return(res);
}
public double FindMin(IFunction fun)
{
List<Molecule> molecules = new List<Molecule>();
double[] bestSwarmPosition = RandGenerator.RandomArray(dimension, Space[0], Space[1]);
for (int i = 0; i < moleculesCount; i++)
{
molecules.Add(
new Molecule(
RandGenerator.RandomArray(dimension, Space[0], Space[1]),
RandGenerator.RandomArray(dimension, -1.0, 1.0))
);
if (fun.Calculate(molecules.Last().bestPosition) < fun.Calculate(bestSwarmPosition))
bestSwarmPosition = molecules.Last().bestPosition;
}
for (int i = 0; i < iterarion; i++)
{
foreach (Molecule p in molecules)
{
for (int j = 0; j < dimension; j++)
{
double randomPositionFactor = random.NextDouble();
double randomSwarmFactor = random.NextDouble();
p.velocity[j] = (p.velocity[j] * velocity) + bestPositionFactor * randomPositionFactor *
(p.bestPosition[j] - p.position[j]) +bestSwarmPos * randomSwarmFactor * (bestSwarmPosition[j] - p.position[j]);
}
for (int j = 0; j < dimension; j++) p.position[j] += p.velocity[j];
if (fun.Calculate(p.position) < fun.Calculate(p.bestPosition)) p.bestPosition = p.position;
if (fun.Calculate(p.position) < fun.Calculate(bestSwarmPosition)) bestSwarmPosition = p.position;
}
}
return fun.Calculate(bestSwarmPosition);
}
public void DrawCommand2(Graphics g)
{
CreateFromConfigs2();
var x1 = double.Parse(X1);
var x2 = double.Parse(X2);
var dx = double.Parse(Dx);
var t = double.Parse(T);
for (double x = x1; x <= x2; x += dx)
{
try
{
var point = GetPointAt(x, _eSumSeries.Calculate(x, t));
g.FillEllipse(new SolidBrush(Color.Black), new Rectangle(point.X - 2, point.Y - 2, 4, 4));
point = GetPointAt(x, _eAnalytical.Calculate(x, t));
g.FillEllipse(new SolidBrush(Color.Red), new Rectangle(point.X - 2, point.Y - 2, 4, 4));
}
catch (Exception ex)
{
Trace.WriteLine($"Drawing point failed: {ex.Message}");
}
}
}
public void testResult(Func <double[], double[]> func)
{
Console.WriteLine("\n\nTest result approximation");
//Shepard model = new Shepard(functionDimension, points);
IFunction model = (IFunction)approx;
for (int i = 1; i < 10; i += 2)
{
double[] a = new double[config.FunctionDimension + config.DependentVariablesNum];
double[] b = new double[config.FunctionDimension + config.DependentVariablesNum];
for (int j = 0; j < config.FunctionDimension; j++)
{
a[j] = model.Min[j] + ((model.Max[j] - model.Min[j]) * i / 10);
b[j] = a[j];
}
double[] results = new double[config.DependentVariablesNum];
for (int j = 0; j < config.DependentVariablesNum; j++)
{
results[j] = a[j + config.FunctionDimension];
}
model.Calculate(a);
Console.WriteLine("Vector " + String.Join(", ", b) + " Approximation result " + String.Join(", ", results) + " real " + String.Join(", ", func(b)) + " error " + approx.GetError(results, func(b)));
}
}
public void CalculateOutput()
{
Output = Type == NeuronType.Input
? Inputs.First()
: Function.Calculate(Inputs.Select((t, i) => t * Weights[i]).Sum() + Bias);
}
public void Calculate(double[] xy)
{
func.Calculate(xy);
}
public IOutBlackBoxParam Function(Double x1, Double x2)
{
calls++;
return(function.Calculate(x1, x2));
}
private double[] build_features(double[] point, IFunction model, Grid grid, double[] distToKnownPoints, double[][] knownPoints = null, int index = -1)
{
// на сколько образующая домен точка близка
// сколько до и после монотонно
// расстояние до известной точки
Analyzer analyzer = new Analyzer(model, knownPoints);
analyzer.do_some_analyse();
// min, max in locality
double maxNeighbours = double.MinValue;
double minNeighbours = double.MaxValue;
foreach (var neighbour in grid.Neighbours(index))
{
double[] calcNeighbour = (double[])grid.Node[neighbour].Clone();
model.Calculate(calcNeighbour);
double calcNeighbourVal = calcVal(model.M, model.N, calcNeighbour);
if (calcNeighbour[calcNeighbour.Length - 1] < minNeighbours)
{
minNeighbours = calcNeighbourVal;
}
if (calcNeighbour[calcNeighbour.Length - 1] > maxNeighbours)
{
maxNeighbours = calcNeighbourVal;
}
}
// current val
double[] curentNode = (double[])grid.Node[index].Clone();
model.Calculate(curentNode);
double curentNodeVal = calcVal(model.M, model.N, curentNode);
if (curentNodeVal < minNeighbours)
{
minNeighbours = curentNodeVal;
}
if (curentNodeVal > maxNeighbours)
{
maxNeighbours = curentNodeVal;
}
List <double[]> temp_points = new List <double[]>();
temp_points = knownPoints.ToList();
temp_points.RemoveAt(analyzer.Domain(grid.Node[index]));
ShepardApprox new_model = new ShepardApprox(model.N, temp_points.ToArray());
double[] old_model_point = (double[])grid.Node[index].Clone();
model.Calculate(old_model_point);
double[] new_model_point = (double[])grid.Node[index].Clone();
new_model.Calculate(new_model_point);
double err = 0;
for (int k = 0; k < model.M; ++k)
{
double newErr = Math.Abs(old_model_point[model.N + k] - new_model_point[model.N + k]);
if (newErr > err)
{
err = newErr;
}
}
this.featureCount = 4;
double[] features = new double[featureCount];
features[0] = maxNeighbours - curentNodeVal;
features[1] = curentNodeVal - minNeighbours;
features[2] = distToKnownPoints[index];
features[3] = err;
return(features);
}