public void ComputeTest()
{
Accord.Math.Tools.SetupGenerator(0);
// Let's create a random dataset containing
// 5000 samples of two dimensional samples.
//
double[,] source = Matrix.Random(5000, 2);
// Now, we will mix the samples the dimensions of the samples.
// A small amount of the second column will be applied to the
// first, and vice-versa.
//
double[,] mix =
{
{ 0.25, 0.25 },
{ -0.25, 0.75 },
};
// mix the source data
double[,] input = source.Multiply(mix);
// Now, we can use ICA to identify any linear mixing between the variables, such
// as the matrix multiplication we did above. After it has identified it, we will
// be able to revert the process, retrieving our original samples again
// Create a new Independent Component Analysis
var ica = new IndependentComponentAnalysis(input);
Assert.AreEqual(IndependentComponentAlgorithm.Parallel, ica.Algorithm);
Assert.AreEqual(ica.Contrast.GetType(), typeof(Logcosh));
// Compute it
ica.Compute();
// Now, we can retrieve the mixing and demixing matrices that were
// used to alter the data. Note that the analysis was able to detect
// this information automatically:
double[][] mixingMatrix = ica.MixingMatrix; // same as the 'mix' matrix
double[][] revertMatrix = ica.DemixingMatrix; // inverse of the 'mix' matrix
double[,] result = ica.Result;
// Verify mixing matrix
mixingMatrix = mixingMatrix.Divide(mixingMatrix.Sum());
Assert.IsTrue(mix.IsEqual(mixingMatrix, atol: 0.008));
// Verify demixing matrix
double[,] expected =
{
{ 0.75, -0.25 },
{ 0.25, 0.25 },
};
revertMatrix = revertMatrix.Divide(revertMatrix.Sum());
Assert.IsTrue(expected.IsEqual(revertMatrix, atol: 0.008));
}
