Changeset 8982 for trunk/sources/HeuristicLab.Tests/HeuristicLab.Algorithms.DataAnalysis-3.4/GaussianProcessFunctionsTest.cs

Timestamp:

12/01/12 19:02:47 (11 years ago)

Author:

gkronber

Message:

#1902: removed class HyperParameter and changed implementations of covariance and mean functions to remove the parameter value caching and event handlers for parameter caching. Instead it is now possible to create the actual covariance and mean functions as Func from templates and specified parameter values. The instances of mean and covariance functions configured in the GUI are actually templates where the structure and fixed parameters can be specified.

File:

• trunk/sources/HeuristicLab.Tests/HeuristicLab.Algorithms.DataAnalysis-3.4/GaussianProcessFunctionsTest.cs (modified) (11 diffs)

trunk/sources/HeuristicLab.Tests/HeuristicLab.Algorithms.DataAnalysis-3.4/GaussianProcessFunctionsTest.cs

@@ -66 +66 @@
           }
       );
+      sum = new MeanSum();
+      sum.Terms.Add(new MeanConst());
+      sum.Terms.Add(new MeanConst());
       TestMeanFunction(sum, 1,
         new double[] { 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 },
@@ -88 +91 @@
           }
       );
+      prod = new MeanProduct();
+      prod.Factors.Add(new MeanConst());
+      prod.Factors.Add(new MeanConst());
       TestMeanFunction(prod, 1,
         new double[] { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 },
@@ -97 +103 @@
       );
 
-      prod.Factors.Clear();
+      prod = new MeanProduct();
       prod.Factors.Add(new MeanZero());
       prod.Factors.Add(new MeanLinear());
@@ -111 +117 @@
           }
       );
+      prod = new MeanProduct();
+      prod.Factors.Add(new MeanZero());
+      prod.Factors.Add(new MeanLinear());
       TestMeanFunction(prod, 1,
         new double[] { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
@@ -123 +132 @@
       );
 
-      prod.Factors.Clear();
+      prod = new MeanProduct();
       prod.Factors.Add(new MeanConst());
       prod.Factors.Add(new MeanLinear());
@@ -1334 +1343 @@
           }
       );
+      cov = new CovarianceSum();
+      cov.Terms.Add(new CovarianceSquaredExponentialIso());
+      cov.Terms.Add(new CovarianceLinear());
       TestCovarianceFunction(cov, 1,
         new double[,]
@@ -1441 +1453 @@
           }
       );
+      cov = new CovarianceScale();
+      cov.CovarianceFunctionParameter.Value = new CovarianceSquaredExponentialIso();
       TestCovarianceFunction(cov, 1,
         new double[,]
@@ -1550 +1564 @@
           }
       );
+      cov = new CovarianceProduct();
+      cov.Factors.Add(new CovarianceSquaredExponentialIso());
+      cov.Factors.Add(new CovarianceLinear());
       TestCovarianceFunction(cov, 1,
         new double[,]
@@ -1603 +1620 @@
       int nHyp = cf.GetNumberOfParameters(x.GetLength(1));
       var hyp = Enumerable.Repeat(hypValue, nHyp).ToArray();
-      cf.SetParameter(hyp);
 
       int rows0 = x.GetLength(0);
       int rows1 = xt.GetLength(0);
       var actualCov = new double[rows0, rows1];
+      var covFunction = cf.GetParameterizedCovarianceFunction(hyp, Enumerable.Range(0, x.GetLength(1)));
       for (int i = 0; i < rows0; i++)
         for (int j = 0; j < rows1; j++)
-          actualCov[i, j] = cf.GetCrossCovariance(x, xt, i, j);
+          actualCov[i, j] = covFunction.CrossCovariance(x, xt, i, j);
 
       AssertEqual(expectedCov, actualCov, delta);
@@ -1616 +1633 @@
       for (int i = 0; i < rows0; i++)
         for (int j = 0; j < rows1; j++) {
-          var g = cf.GetGradient(x, i, j).ToArray();
+          var g = covFunction.CovarianceGradient(x, i, j).ToArray();
           for (int k = 0; k < nHyp; k++)
             Assert.AreEqual(expectedGradients[k][i, j], g[k], delta);
@@ -1629 +1646 @@
       int nHyp = mf.GetNumberOfParameters(x.GetLength(1));
       var hyp = Enumerable.Repeat(hypValue, nHyp).ToArray();
-      mf.SetParameter(hyp);
-
-      var m = mf.GetMean(xt);
+      var meanFunction = mf.GetParameterizedMeanFunction(hyp, Enumerable.Range(0, x.GetLength(1)));
+
+      var m = Enumerable.Range(0, xt.GetLength(0)).Select(i => meanFunction.Mean(xt, i)).ToArray();
 
       AssertEqual(expectedMean, m);
 
-      for (int i = 0; i < nHyp; i++) {
-        var g = mf.GetGradients(i, x);
-        AssertEqual(expectedGradients[i], g);
+      for (int k = 0; k < nHyp; k++) {
+        var g = Enumerable.Range(0, xt.GetLength(0)).Select(i => meanFunction.Gradient(x, i, k)).ToArray();
+        AssertEqual(expectedGradients[k], g);
       }
     }