public void MiniBatchKMeansMoreWeightsThanSamples()
{
Accord.Math.Tools.SetupGenerator(0);
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
double[] weights = { 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0 };
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
Assert.Throws <DimensionMismatchException>(() => mbkmeans.Learn(observations, weights), "");
}
public void MiniBatchKMeansNoSamples()
{
Accord.Math.Tools.SetupGenerator(0);
// Declare some observations
double[][] observations = null;
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
Assert.Throws <ArgumentNullException>(() => mbkmeans.Learn(observations), "");
}
public void MiniBatchKMeansConstructorTest()
{
Accord.Math.Random.Generator.Seed = 0;
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
double[][] orig = observations.MemberwiseClone();
// Create a new Mini-Batch K-Means algorithm with 3 clusters
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
// Compute the algorithm, retrieving an integer array
// containing the labels for each of the observations
int[] labels = mbkmeans.Learn(observations).Decide(observations);
// As a result, the first two observations should belong to the
// same cluster (thus having the same label). The same should
// happen to the next four observations and to the last three.
Assert.AreEqual(labels[0], labels[1]);
Assert.AreEqual(labels[2], labels[3]);
Assert.AreEqual(labels[2], labels[4]);
Assert.AreEqual(labels[2], labels[5]);
Assert.AreEqual(labels[6], labels[7]);
Assert.AreEqual(labels[6], labels[8]);
Assert.AreNotEqual(labels[0], labels[2]);
Assert.AreNotEqual(labels[2], labels[6]);
Assert.AreNotEqual(labels[0], labels[6]);
int[] labels2 = mbkmeans.Clusters.Decide(observations);
Assert.IsTrue(labels.IsEqual(labels2));
// the data must not have changed!
Assert.IsTrue(orig.IsEqual(observations));
}
public void uniform_sampling_test()
{
Accord.Math.Random.Generator.Seed = 0;
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
// Create a new K-Means algorithm
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(k: 3, batchSize: 2)
{
UseSeeding = Seeding.Uniform
};
// Compute and retrieve the data centroids
var clusters = mbkmeans.Learn(observations);
int[] labels = clusters.Decide(observations);
int[] labels2 = mbkmeans.Clusters.Decide(observations);
// the data must not have changed!
double[][] orig =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
Assert.IsTrue(orig.IsEqual(observations));
}
public void MiniBatchKMeansNotRectangularData()
{
Accord.Math.Tools.SetupGenerator(0);
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1,2, 0 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
Assert.Throws <DimensionMismatchException>(() => mbkmeans.Learn(observations), "");
}
public void MiniBatchKMeansMorePointsInBatchThanSamples()
{
Accord.Math.Tools.SetupGenerator(0);
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 15);
Assert.Throws <ArgumentException>(() => mbkmeans.Learn(observations), "");
}
public void MiniBatchKMeansNegativeSumOfWeights()
{
Accord.Math.Tools.SetupGenerator(0);
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
double[] weights = new double[] { -1, -1, -1, -1, -1, -1, -1, -1, -1 };
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
Assert.Throws <ArgumentException>(() => mbkmeans.Learn(observations, weights), "");
}
public void learn_test()
{
#region doc_learn
Accord.Math.Random.Generator.Seed = 0;
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
// Create a new Mini-Batch K-Means algorithm
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(k: 3, batchSize: 2);
// Compute and retrieve the data centroids
var clusters = mbkmeans.Learn(observations);
// Use the centroids to parition all the data
int[] labels = clusters.Decide(observations);
#endregion
Assert.AreEqual(labels[0], labels[1]);
Assert.AreEqual(labels[2], labels[3]);
Assert.AreEqual(labels[2], labels[4]);
Assert.AreEqual(labels[2], labels[5]);
Assert.AreEqual(labels[6], labels[7]);
Assert.AreEqual(labels[6], labels[8]);
Assert.AreNotEqual(labels[0], labels[2]);
Assert.AreNotEqual(labels[2], labels[6]);
Assert.AreNotEqual(labels[0], labels[6]);
int[] labels2 = mbkmeans.Clusters.Decide(observations);
Assert.IsTrue(labels.IsEqual(labels2));
// the data must not have changed!
double[][] orig =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
Assert.IsTrue(orig.IsEqual(observations));
var c = new KMeansClusterCollection.KMeansCluster[clusters.Count];
int i = 0;
foreach (var cluster in clusters)
{
c[i++] = cluster;
}
for (i = 0; i < c.Length; i++)
{
Assert.AreSame(c[i], clusters[i]);
}
}
public void MiniBatchKMeansRandomizationTest()
{
Generator.Seed = 1;
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
double error, e;
// Create a new algorithm
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(3, 2);
mbkmeans.Randomize(observations);
// Save the first initialization
double[][] initial = mbkmeans.Clusters.Centroids.MemberwiseClone();
// Compute the first K-Means
// The seed for the random number generator has to be set just before any execution of Learn
// so that the learning methods always uses the same subsequence of the generator's sequence of pseudo-random numbers
// if we want the algorithm to behave the same.
Generator.Seed = 1;
mbkmeans.ComputeError = true;
mbkmeans.Learn(observations);
error = mbkmeans.Error;
// Create more K-Means algorithms
// with the same initializations
for (int i = 0; i < 1000; i++)
{
mbkmeans = new MiniBatchKMeans(3, 2);
mbkmeans.Clusters.Centroids = initial;
mbkmeans.ComputeError = true;
Generator.Seed = 1;
mbkmeans.Learn(observations);
e = mbkmeans.Error;
Assert.AreEqual(error, e);
}
// Create more K-Means algorithms
// without the same initialization
// Now, because we want different randomization
// we need to use different seed. It is sufficient to
// set it only once because different calls to Learn
// will be using different subsequences of pseudo-random numbers in the generator's sequence.
Generator.Seed = 2;
bool differ = false;
for (int i = 0; i < 1000; i++)
{
mbkmeans = new MiniBatchKMeans(3, 2);
mbkmeans.ComputeError = true;
mbkmeans.Learn(observations);
e = mbkmeans.Error;
if (error != e)
{
differ = true;
}
}
Assert.IsTrue(differ);
}
public void learn_test_weights()
{
#region doc_learn_weights
Accord.Math.Random.Generator.Seed = 0;
// A common desire when doing clustering is to attempt to find how to
// weight the different components / columns of a dataset, giving them
// different importances depending on the end goal of the clustering task.
// Declare some observations
double[][] observations =
{
new double[] { -5, -2, -1 },
new double[] { -5, -5, -6 },
new double[] { 2, 1, 1 },
new double[] { 1, 1, 2 },
new double[] { 1, 2, 2 },
new double[] { 3, 1, 2 },
new double[] { 11, 5, 4 },
new double[] { 15, 5, 6 },
new double[] { 10, 5, 6 },
};
// Create a new K-Means algorithm
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(k: 3, batchSize: 2)
{
// For example, let's say we would like to consider the importance of
// the first column as 0.1, the second column as 0.7 and the third 0.9
Distance = new WeightedSquareEuclidean(new double[] { 0.1, 0.7, 1.1 })
};
// Compute and retrieve the data centroids
var clusters = mbkmeans.Learn(observations);
// Use the centroids to parition all the data
int[] labels = clusters.Decide(observations);
#endregion
Assert.AreEqual(labels[0], labels[2]);
Assert.AreEqual(labels[0], labels[2]);
Assert.AreEqual(labels[0], labels[3]);
Assert.AreEqual(labels[0], labels[4]);
Assert.AreEqual(labels[0], labels[4]);
Assert.AreEqual(labels[6], labels[7]);
Assert.AreEqual(labels[6], labels[8]);
Assert.AreNotEqual(labels[0], labels[1]);
Assert.AreNotEqual(labels[2], labels[6]);
Assert.AreNotEqual(labels[0], labels[6]);
int[] labels2 = mbkmeans.Clusters.Decide(observations);
Assert.IsTrue(labels.IsEqual(labels2));
var c = new KMeansClusterCollection.KMeansCluster[clusters.Count];
int i = 0;
foreach (var cluster in clusters)
{
c[i++] = cluster;
}
for (i = 0; i < c.Length; i++)
{
Assert.AreSame(c[i], clusters[i]);
}
}
public void learn_test_mixed()
{
#region doc_learn_mixed
Accord.Math.Random.Generator.Seed = 0;
// Declare some mixed discrete and continuous observations
double[][] observations =
{
// (categorical) (discrete) (continuous)
new double[] { 1, -1, -2.2 },
new double[] { 1, -6, -5.5 },
new double[] { 2, 1, 1.1 },
new double[] { 2, 2, 1.2 },
new double[] { 2, 2, 2.6 },
new double[] { 3, 2, 1.4 },
new double[] { 3, 4, 5.2 },
new double[] { 1, 6, 5.1 },
new double[] { 1, 6, 5.9 },
};
// Create a new codification algorithm to convert
// the mixed variables above into all continuous:
var codification = new Codification <double>()
{
CodificationVariable.Categorical,
CodificationVariable.Discrete,
CodificationVariable.Continuous
};
// Learn the codification from observations
var model = codification.Learn(observations);
// Transform the mixed observations into only continuous:
double[][] newObservations = model.ToDouble().Transform(observations);
// (newObservations will be equivalent to)
double[][] expected =
{
// (one hot) (discrete) (continuous)
new double[] { 1, 0, 0, -1, -2.2 },
new double[] { 1, 0, 0, -6, -5.5 },
new double[] { 0, 1, 0, 1, 1.1 },
new double[] { 0, 1, 0, 2, 1.2 },
new double[] { 0, 1, 0, 2, 2.6 },
new double[] { 0, 0, 1, 2, 1.4 },
new double[] { 0, 0, 1, 4, 5.2 },
new double[] { 1, 0, 0, 6, 5.1 },
new double[] { 1, 0, 0, 6, 5.9 },
};
// Create a new K-Means algorithm
MiniBatchKMeans mbkmeans = new MiniBatchKMeans(k: 3, batchSize: 2);
// Compute and retrieve the data centroids
var clusters = mbkmeans.Learn(observations);
// Use the centroids to parition all the data
int[] labels = clusters.Decide(observations);
#endregion
Assert.IsTrue(expected.IsEqual(newObservations, 1e-8));
Assert.AreEqual(3, codification.NumberOfInputs);
Assert.AreEqual(5, codification.NumberOfOutputs);
Assert.AreEqual(3, codification.Columns.Count);
Assert.AreEqual("0", codification.Columns[0].ColumnName);
Assert.AreEqual(3, codification.Columns[0].NumberOfSymbols);
Assert.AreEqual(1, codification.Columns[0].NumberOfInputs);
Assert.AreEqual(3, codification.Columns[0].NumberOfOutputs);
Assert.AreEqual(3, codification.Columns[0].NumberOfClasses);
Assert.AreEqual(CodificationVariable.Categorical, codification.Columns[0].VariableType);
Assert.AreEqual("1", codification.Columns[1].ColumnName);
Assert.AreEqual(1, codification.Columns[1].NumberOfSymbols);
Assert.AreEqual(1, codification.Columns[1].NumberOfInputs);
Assert.AreEqual(1, codification.Columns[1].NumberOfOutputs);
Assert.AreEqual(1, codification.Columns[1].NumberOfClasses);
Assert.AreEqual(CodificationVariable.Discrete, codification.Columns[1].VariableType);
Assert.AreEqual("2", codification.Columns[2].ColumnName);
Assert.AreEqual(1, codification.Columns[2].NumberOfSymbols);
Assert.AreEqual(1, codification.Columns[2].NumberOfInputs);
Assert.AreEqual(1, codification.Columns[2].NumberOfOutputs);
Assert.AreEqual(1, codification.Columns[2].NumberOfClasses);
Assert.AreEqual(CodificationVariable.Continuous, codification.Columns[2].VariableType);
Assert.AreEqual(labels[0], labels[2]);
Assert.AreEqual(labels[0], labels[3]);
Assert.AreEqual(labels[0], labels[4]);
Assert.AreEqual(labels[0], labels[5]);
Assert.AreEqual(labels[6], labels[7]);
Assert.AreEqual(labels[6], labels[8]);
Assert.AreNotEqual(labels[0], labels[1]);
Assert.AreNotEqual(labels[0], labels[6]);
int[] labels2 = mbkmeans.Clusters.Decide(observations);
Assert.IsTrue(labels.IsEqual(labels2));
var c = new KMeansClusterCollection.KMeansCluster[clusters.Count];
int i = 0;
foreach (var cluster in clusters)
{
c[i++] = cluster;
}
for (i = 0; i < c.Length; i++)
{
Assert.AreSame(c[i], clusters[i]);
}
}