public void GaussianProcessRegressionPerformanceTest() {
ex = null;
var alg = new GaussianProcessRegression();
alg.Engine = new HeuristicLab.SequentialEngine.SequentialEngine();
alg.Problem = new RegressionProblem();
var provider = new RegressionCSVInstanceProvider();
var problemData = (RegressionProblemData)provider.ImportData(@"Test Resources\co2.txt");
problemData.TargetVariableParameter.ActualValue = problemData.TargetVariableParameter.ValidValues.First(x => x.Value == "interpolated");
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "year"), false);
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "month"), false);
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "average"), false);
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "interpolated"), false);
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "trend"), false);
problemData.InputVariables.SetItemCheckedState(problemData.InputVariables.First(x => x.Value == "#days"), false);
alg.Problem.ProblemDataParameter.Value = problemData;
alg.ExceptionOccurred += new EventHandler<EventArgs<Exception>>(cv_ExceptionOccurred);
alg.Stopped += new EventHandler(cv_Stopped);
alg.Prepare();
alg.Start();
trigger.WaitOne();
if (ex != null) throw ex;
TestContext.WriteLine("Runtime: {0}", alg.ExecutionTime.ToString());
}
public void GaussianProcessModelOutputTest() {
var provider = new RegressionCSVInstanceProvider();
var problemData = provider.ImportData(@"Test Resources\co2.txt");
var targetVariable = "interpolated";
var allowedInputVariables = new string[] { "decimal date" };
var rows = Enumerable.Range(0, 401);
var meanFunction = new MeanConst();
var covarianceFunction = new CovarianceSum();
covarianceFunction.Terms.Add(new CovarianceSquaredExponentialIso());
var prod = new CovarianceProduct();
prod.Factors.Add(new CovarianceSquaredExponentialIso());
prod.Factors.Add(new CovariancePeriodic());
covarianceFunction.Terms.Add(prod);
{
var hyp = new double[] { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
var model = new GaussianProcessModel(problemData.Dataset, targetVariable, allowedInputVariables, rows, hyp,
meanFunction,
covarianceFunction);
Assert.AreEqual(4.3170e+004, model.NegativeLogLikelihood, 1);
var dHyp = model.HyperparameterGradients;
Assert.AreEqual(-248.7932, dHyp[0], 1E-2);
var dHypCovExpected = new double[] { -0.5550e4, -5.5533e4, -0.2511e4, -2.7625e4, -1.3033e4, 0.0289e4, -2.7625e4 };
AssertEqual(dHypCovExpected, dHyp.Skip(1).Take(7).ToArray(), 1);
Assert.AreEqual(-2.0171e+003, dHyp.Last(), 1);
var predTrain = model.GetEstimatedValues(problemData.Dataset, new int[] { 0, 400 }).ToArray();
Assert.AreEqual(310.5930, predTrain[0], 1e-3);
Assert.AreEqual(347.9993, predTrain[1], 1e-3);
var predTrainVar = model.GetEstimatedVariance(problemData.Dataset, problemData.TrainingIndices).ToArray();
}
{
var hyp = new double[] { 0.029973094285941, 0.455535210579926, 3.438647883940457, 1.464114485889487, 3.001788584487478, 3.815289323309630, 4.374914122810222, 3.001788584487478, 0.716427415979145 };
var model = new GaussianProcessModel(problemData.Dataset, targetVariable, allowedInputVariables, rows, hyp,
meanFunction,
covarianceFunction);
Assert.AreEqual(872.8448, model.NegativeLogLikelihood, 1e-3);
var dHyp = model.HyperparameterGradients;
Assert.AreEqual(-0.0046, dHyp[0], 1e-3);
var dHypCovExpected = new double[] { 0.2652, -0.2386, 0.1706, -0.1744, 0.0000, 0.0000, -0.1744 };
AssertEqual(dHypCovExpected, dHyp.Skip(1).Take(7).ToArray(), 1e-3);
Assert.AreEqual(0.8621, dHyp.Last(), 1e-3);
var predTrain = model.GetEstimatedValues(problemData.Dataset, new int[] { 0, 400 }).ToArray();
Assert.AreEqual(315.3692, predTrain[0], 1e-3);
Assert.AreEqual(356.6076, predTrain[1], 1e-3);
}
}
