public void Run()
{
var func = new MyObjecticeFunctionForSA();
var opt = new LibOptimization.Optimization.clsOptSimulatedAnnealing(func);
//initial position using random
var rng = new LibOptimization.Util.clsRandomXorshift((UInt32)DateTime.Now.Millisecond);
opt.InitialPosition = new double[] { rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2) };
//parameter for SA
opt.Temperature = 1;
opt.StopTemperature = 0.0001;
opt.CoolingRatio = 0.9995;
opt.IsUseCriterion = false;
opt.Iteration = 50000;
//neighbor function
{
//default neightbor function(default)
//var myNeighborFunc = new LocalRandomSearch();
//myNeighborFunc.NeighborRange = 0.0001;
//opt.Neighbor = myNeighborFunc;
}
{
//designed neightbor function
var myNeighborFunc = new MyNeighborFunc();
myNeighborFunc.Range = 1;
opt.Neighbor = myNeighborFunc;
}
//init
opt.Init();
clsUtil.DebugValue(opt);
//do optimization
while (opt.DoIteration(2000) == false)
{
clsUtil.DebugValue(opt);
}
clsUtil.DebugValue(opt);
//restart
opt.Init(opt.Result);
clsUtil.DebugValue(opt);
//do optimization
while (opt.DoIteration(2000) == false)
{
clsUtil.DebugValue(opt);
}
clsUtil.DebugValue(opt);
}
static void Main(string[] args)
{
//Simulated Annealing
{
var func = new SphereFunction();
var opt = new LibOptimization.Optimization.clsOptSimulatedAnnealing(func);
//initial position using random
var rng = new LibOptimization.Util.clsRandomXorshift((UInt32)DateTime.Now.Millisecond);
opt.InitialPosition = new double[] { rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2), rng.NextDouble(-2, 2) };
//parameter for SA
opt.Temperature = 1;
opt.StopTemperature = 0.0001;
opt.CoolingRatio = 0.99;
opt.Iteration = 30000; //default 20000
//neighbor
((LocalRandomSearch)opt.Neighbor).NeighborRange = 0.001; //neghbor function
//init
opt.Init();
clsUtil.DebugValue(opt);
//do optimization
while (opt.DoIteration(2000) == false)
{
clsUtil.DebugValue(opt);
//my criterion
/*
* if (Result.Eval < 0.01)
* {
* break;
* }
* else
* {
* Console.WriteLine("Eval:{0}", opt.Result.Eval);
* }
*/
}
clsUtil.DebugValue(opt);
}
//Typical use
{
//Target Function
var func = new RosenBrock(2);
//Set Function
var opt = new LibOptimization.Optimization.clsOptNelderMead(func);
opt.Init();
//Optimization
opt.DoIteration();
//Check Error
if (opt.IsRecentError() == true)
{
return;
}
else
{
//Get Result
clsUtil.DebugValue(opt);
}
}
//Evaluate optimization result per 100 iteration
{
var opt = new LibOptimization.Optimization.clsOptDEJADE(new RosenBrock(10));
opt.Init();
clsUtil.DebugValue(opt);
while (opt.DoIteration(100) == false)
{
clsUtil.DebugValue(opt, ai_isOutValue: false);
}
clsUtil.DebugValue(opt);
}
//Evaluate optimization result per 100 iteration with check my criterion.
{
var opt = new LibOptimization.Optimization.clsOptDEJADE(new RosenBrock(10));
//Disable Internal criterion
opt.IsUseCriterion = false;
//Init
opt.Init();
clsUtil.DebugValue(opt);
//do optimization!
while (opt.DoIteration(100) == false)
{
var eval = opt.Result.Eval;
//my criterion
if (eval < 0.01)
{
break;
}
else
{
clsUtil.DebugValue(opt, ai_isOutValue: false);
}
}
clsUtil.DebugValue(opt);
}
//Set boundary variable.
//-20<x1<-1, -15<x2<0
{
var opt = new LibOptimization.Optimization.clsOptDEJADE(new RosenBrock(2));
//move initial position
opt.InitialPosition = new double[] { -10, -10 };
//Set boundary variable
opt.LowerBounds = new double[] { -20, -15 };
opt.UpperBounds = new double[] { -1, 0 };
//Init
opt.Init();
clsUtil.DebugValue(opt);
//do optimization!
while (opt.DoIteration(100) == false)
{
var eval = opt.Result.Eval;
//my criterion
if (eval < 0.01)
{
break;
}
else
{
clsUtil.DebugValue(opt, ai_isOutValue: false);
}
}
clsUtil.DebugValue(opt);
}
//Optimiztion problem using MyObjectiveFunction
// min f(x)
// s.t. x>0, 170<=x1<=200, 200<=x2<=300, 250<=x3<=400, 370<=x4<=580, 380<=x5<=600
{
var func = new MyObjectiveFunction();
var opt = new LibOptimization.Optimization.clsOptDEJADE(func);
//Set boundary variable
opt.LowerBounds = new double[] { 170, 200, 250, 370, 380 };
opt.UpperBounds = new double[] { 200, 300, 400, 580, 600 };
//move initial position
double[] initialPosition = new double[] { 0, 0, 0, 0, 0 };
for (int i = 0; i < initialPosition.Length; i++)
{
//center of boundary range.
initialPosition[i] = (opt.LowerBounds[i] + opt.UpperBounds[i]) / 2.0;
}
opt.InitialPosition = initialPosition;
//Init
opt.Init();
clsUtil.DebugValue(opt);
//do optimization!
while (opt.DoIteration(100) == false)
{
var eval = opt.Result.Eval;
//my criterion
if (eval < 0.01)
{
break;
}
else
{
Console.WriteLine("Eval:{0}", opt.Result.Eval);
}
}
clsUtil.DebugValue(opt);
//return;
}
}
