public static void Train(double[][] dataMatrix)
{
Gaussian kernel = new Gaussian(10.0);
var kpca = new KernelPrincipalComponentAnalysis(kernel);
model = kpca.Learn(dataMatrix);
}
public void transform_more_columns_than_samples_new_interface()
{
// Lindsay's tutorial data
double[,] datat = data.Transpose();
var target = new KernelPrincipalComponentAnalysis(new Linear());
// Compute
target.Learn(datat);
// Transform
double[,] actual = target.Transform(datat);
// Assert the scores equals the transformation of the input
double[,] result = target.Result;
double[,] expected = new double[, ]
{
{ 0.50497524691810358 },
{ -0.504975246918104 }
}.Multiply(-1);
Assert.IsTrue(Matrix.IsEqual(expected, actual, 0.01));
Assert.IsTrue(Matrix.IsEqual(result, actual, 0.01));
}
// GET: DataAnalysis/PCA_Analysis/id
public async Task <IActionResult> PcaAnalysis(int?id)
{
if (id == null)
{
return(NotFound());
}
var dataAnalysis = await _context.FilesInformation.FindAsync(id);
if (dataAnalysis == null)
{
return(NotFound());
}
string path = await _context.FilesInformation.Where(m => m.Id == id).Select(d => d.Path).FirstOrDefaultAsync();
CsvReader reader = new CsvReader(_appEnvironment.WebRootPath + path, hasHeaders: true);
double[][] actual = reader.ToJagged();
var pcaTool = new KernelPrincipalComponentAnalysis();
pcaTool.Learn(actual);
pcaTool.NumberOfOutputs = 3;
var outputMatrix = pcaTool.Transform(actual);
ViewBag.columns = pcaTool.NumberOfOutputs;
ViewBag.rows = outputMatrix.Length;
List <double> PC1 = new List <double>();
List <double> PC2 = new List <double>();
List <double> PC3 = new List <double>();
for (int i = 0; i < outputMatrix.Length; i++)
{
for (int j = 0; j < pcaTool.NumberOfOutputs; j += 3)
{
outputMatrix[i][j] = Math.Round(outputMatrix[i][j], 3);
outputMatrix[i][j + 1] = Math.Round(outputMatrix[i][j + 1], 3);
outputMatrix[i][j + 2] = Math.Round(outputMatrix[i][j + 2], 3);
PC1.Add(outputMatrix[i][j]);
PC2.Add(outputMatrix[i][j + 1]);
PC3.Add(outputMatrix[i][j + 2]);
}
}
ViewBag.data = outputMatrix;
ViewBag.PC1 = PC1;
ViewBag.PC2 = PC2;
ViewBag.PC3 = PC3;
return(View());
}
public void learn_whiten_success()
{
double[,] data =
{
{ 2.5, 2.4 },
{ 0.5, 0.7 },
{ 2.2, 2.9 },
{ 1.9, 2.2 },
{ 3.1, 3.0 },
{ 2.3, 2.7 },
{ 2.0, 1.6 },
{ 1.0, 1.1 },
{ 1.5, 1.6 },
{ 1.1, 0.9 }
};
var method = PrincipalComponentMethod.Center; // PrincipalComponentMethod.Standardize
var pca = new KernelPrincipalComponentAnalysis(new Linear(), method, whiten: true);
pca.Learn(data);
double[] eigenvalues = { 1.28402771, 0.0490833989 };
double[] proportion = eigenvalues.Divide(eigenvalues.Sum());
double[,] eigenvectors =
{
{ 0.19940687993951403, -1.1061252858739095 },
{ 0.21626410214440508, 1.0199057073792104 }
};
// Everything is alright (up to the 9 decimal places shown in the tutorial)
// Assert.IsTrue(eigenvectors.IsEqual(pca.ComponentMatrix, rtol: 1e-9));
Assert.IsTrue(proportion.IsEqual(pca.ComponentProportions, rtol: 1e-9));
Assert.IsTrue(eigenvalues.IsEqual(pca.Eigenvalues.Divide(data.GetLength(0) - 1), rtol: 1e-5));
double[,] actual = pca.Transform(data);
double[][] expected = new double[][]
{
new double[] { 0.243560157209023, -0.263472650637184 },
new double[] { -0.522902576315494, 0.214938218565977 },
new double[] { 0.291870144299372, 0.578317788814594 },
new double[] { 0.0806632088164338, 0.19622137941132 },
new double[] { 0.492962746459375, -0.315204397734004 },
new double[] { 0.268558011864442, 0.263724118751361 },
new double[] { -0.0291545644762578, -0.526334573603598 },
new double[] { -0.336693495487974, 0.0698378585807067 },
new double[] { -0.128858004446015, 0.0267280693333571 },
new double[] { -0.360005627922904, -0.244755811482527 }
}.Multiply(-1);
// Everything is correct (up to 8 decimal places)
Assert.IsTrue(expected.IsEqual(actual, atol: 1e-8));
}
public void test_kernlab_new_method()
{
// Tested against R's kernlab
// test <- c(16, 117, 94, 132, 13, 73, 68, 129, 91, 50, 56, 12, 145, 105, 35, 53, 38, 51, 85, 116)
int[] test = { 15, 116, 93, 131, 12, 72, 67, 128, 90, 49, 55, 11, 144, 104, 34, 52, 37, 50, 84, 115 };
// data(iris)
double[,] iris = create_iris();
// kpc <- kpca(~.,data=iris[-test,-5],kernel="rbfdot",kpar=list(sigma=0.2),features=2)
var data = iris.Remove(test, new[] { 4 });
var kernel = new Gaussian()
{
Gamma = 1
};
var kpc = new KernelPrincipalComponentAnalysis(kernel);
kpc.NumberOfOutputs = 2;
var transform = kpc.Learn(data);
var rotated = kpc.Result;
var pcv = kpc.ComponentMatrix;
var eig = kpc.Eigenvalues;
double[] expected_eig = { 28.542404060412132, 15.235596653653861 };
double[,] expected_pcv = expected_r();
Assert.IsTrue(Matrix.IsEqual(expected_eig, eig, 1e-10));
Assert.IsTrue(Matrix.IsEqual(expected_pcv, pcv, 1e-10));
double[,] irisSubset = iris_sub();
var testing = iris.Submatrix(test, new[] { 0, 1, 2, 3 });
Assert.IsTrue(Matrix.IsEqual(irisSubset, testing));
double[,] expectedProjection = expected_p();
double[,] proj = kpc.Transform(testing);
Assert.IsTrue(expectedProjection.IsEqual(proj, 1e-6));
double[][] proj2 = kpc.Transform(testing.ToJagged());
Assert.IsTrue(expectedProjection.IsEqual(proj, 1e-6));
double[][] proj3 = transform.Transform(testing.ToJagged());
Assert.IsTrue(expectedProjection.IsEqual(proj, 1e-6));
}
public void SerializeTest()
{
double[][] actual, expected = new double[][]
{
new double[] { -0.57497881446526, 0.0385634996866008 },
new double[] { 0.615576484818799, 0.463702189462175 },
new double[] { -0.593829949403244, 0.225530142279202 },
new double[] { -0.317225395167446, -0.470473936541537 },
new double[] { -0.372910594880684, 0.490761790918429 },
new double[] { -0.635435167456034, 0.154118097007375 },
new double[] { 0.0591892009931548, -0.659772491267272 },
new double[] { 0.739020019118434, 0.133927036952283 },
new double[] { 0.344183901764118, -0.559832164574368 },
new double[] { 0.736410314678163, 0.183475836077113 }
};
var target = new KernelPrincipalComponentAnalysis()
{
Kernel = new Gaussian(0.7)
};
target.Learn(data.ToJagged());
actual = target.Transform(data.ToJagged());
string str = actual.ToCSharp();
Assert.IsTrue(Matrix.IsEqual(actual, expected, 1e-10));
var copy = Serializer.DeepClone(target);
actual = copy.Transform(data.ToJagged());
Assert.IsTrue(Matrix.IsEqual(actual, expected, 1e-10));
Assert.IsTrue(target.Kernel.Equals(copy.Kernel));
Assert.IsTrue(target.ComponentProportions.IsEqual(copy.ComponentProportions));
Assert.IsTrue(target.ComponentVectors.IsEqual(copy.ComponentVectors));
Assert.IsTrue(target.CumulativeProportions.IsEqual(copy.CumulativeProportions));
Assert.IsTrue(target.Eigenvalues.IsEqual(copy.Eigenvalues));
Assert.IsTrue(target.MaximumNumberOfOutputs.IsEqual(copy.MaximumNumberOfOutputs));
Assert.IsTrue(target.Method.Equals(copy.Method));
Assert.IsTrue(target.NumberOfInputs.IsEqual(copy.NumberOfInputs));
Assert.IsTrue(target.NumberOfOutputs.IsEqual(copy.NumberOfOutputs));
Assert.IsTrue(target.Overwrite.Equals(copy.Overwrite));
Assert.IsTrue(target.Whiten.Equals(copy.Whiten));
}
public void kernel_matrix_success()
{
var actual = new KernelPrincipalComponentAnalysis(new Linear(), PrincipalComponentMethod.KernelMatrix);
var expected = new KernelPrincipalComponentAnalysis(data, new Linear(), AnalysisMethod.Standardize);
Linear kernel = new Linear();
double[][] K = kernel.ToJagged(data.ZScores().ToJagged());
// Compute
actual.Learn(K);
expected.Compute();
// Transform
double[][] actualTransform = actual.Transform(K);
double[][] expectedTransform1 = expected.Transform(data).ToJagged();
double[][] expectedTransform2 = expected.Transform(data.ToJagged());
// Verify both are equal with 0.01 tolerance value
Assert.IsTrue(Matrix.IsEqual(actualTransform, expectedTransform1, 0.01));
Assert.IsTrue(Matrix.IsEqual(actualTransform, expectedTransform2, 0.01));
}
public void RevertTest2_new_method()
{
string path = @"Resources\examples.xls";
// Create a new reader, opening a given path
ExcelReader reader = new ExcelReader(path);
// Afterwards, we can query the file for all
// worksheets within the specified workbook:
string[] sheets = reader.GetWorksheetList();
// Finally, we can request an specific sheet:
DataTable table = reader.GetWorksheet("Wikipedia");
// Now, we have loaded the Excel file into a DataTable. We
// can go further and transform it into a matrix to start
// running other algorithms on it:
double[][] matrix = table.ToArray();
IKernel kernel = new Gaussian(5);
// Create analysis
var target = new KernelPrincipalComponentAnalysis(kernel)
{
Method = PrincipalComponentMethod.Center,
Center = true // Center in feature space
};
var regression = target.Learn(matrix);
double[][] forward = regression.Transform(matrix);
double[][] reversion = target.Revert(forward);
Assert.IsTrue(!reversion.HasNaN());
}
/// <summary>
/// Launched when the user clicks the "Run analysis" button.
/// </summary>
///
private void btnCompute_Click(object sender, EventArgs e)
{
// Save any pending changes
dgvAnalysisSource.EndEdit();
if (dgvAnalysisSource.DataSource == null)
{
MessageBox.Show("Please load some data using File > Open!");
return;
}
// Create a matrix from the source data table
double[][] sourceMatrix = (dgvAnalysisSource.DataSource as DataTable).ToJagged(out columnNames);
// Create and compute a new Simple Descriptive Analysis
sda = new DescriptiveAnalysis()
{
ColumnNames = columnNames
};
sda.Learn(sourceMatrix);
// Show the descriptive analysis on the screen
dgvDistributionMeasures.DataSource = sda.Measures;
// Create the kernel function
IKernel kernel = createKernel();
// Get the input values (the two first columns)
this.inputs = sourceMatrix.GetColumns(0, 1);
// Get only the associated labels (last column)
this.outputs = sourceMatrix.GetColumn(2).ToMulticlass();
var method = (PrincipalComponentMethod)cbMethod.SelectedValue;
// Creates the Kernel Principal Component Analysis of the given source
kpca = new KernelPrincipalComponentAnalysis()
{
Kernel = kernel,
Method = method
};
// Whether to center in space
kpca.Center = cbCenter.Checked;
var classifier = kpca.Learn(inputs); // Finally, compute the analysis!
double[][] result = kpca.Transform(inputs);
double[][] reduced;
if (kpca.Components.Count >= 2)
{
// Perform the transformation of the data using two components
kpca.NumberOfOutputs = 2;
reduced = kpca.Transform(inputs);
}
else
{
kpca.NumberOfOutputs = 1;
reduced = kpca.Transform(inputs);
reduced = reduced.InsertColumn(Vector.Zeros(reduced.GetLength(0)));
}
// Create a new plot with the original Z column
double[][] points = reduced.InsertColumn(sourceMatrix.GetColumn(2));
// Create output scatter plot
outputScatterplot.DataSource = points;
CreateScatterplot(graphMapFeature, points);
// Create output table
dgvProjectionResult.DataSource = new ArrayDataView(points, columnNames);
dgvReversionSource.DataSource = new ArrayDataView(result);
// Populates components overview with analysis data
dgvFeatureVectors.DataSource = new ArrayDataView(kpca.ComponentVectors);
dgvPrincipalComponents.DataSource = kpca.Components;
cumulativeView.DataSource = kpca.Components;
distributionView.DataSource = kpca.Components;
numNeighbor.Maximum = result.Rows();
numNeighbor.Value = System.Math.Min(10, numNeighbor.Maximum);
lbStatus.Text = "Good! Feel free to browse the other tabs to see what has been found.";
}
public void learn_success()
{
#region doc_learn_1
// Reproducing Lindsay Smith's "Tutorial on Principal Component Analysis"
// using the framework's default method. The tutorial can be found online
// at http://www.sccg.sk/~haladova/principal_components.pdf
// Step 1. Get some data
// ---------------------
double[][] data =
{
new double[] { 2.5, 2.4 },
new double[] { 0.5, 0.7 },
new double[] { 2.2, 2.9 },
new double[] { 1.9, 2.2 },
new double[] { 3.1, 3.0 },
new double[] { 2.3, 2.7 },
new double[] { 2.0, 1.6 },
new double[] { 1.0, 1.1 },
new double[] { 1.5, 1.6 },
new double[] { 1.1, 0.9 }
};
// Step 2. Subtract the mean
// -------------------------
// Note: The framework does this automatically. By default, the framework
// uses the "Center" method, which only subtracts the mean. However, it is
// also possible to remove the mean *and* divide by the standard deviation
// (thus performing the correlation method) by specifying "Standardize"
// instead of "Center" as the AnalysisMethod.
var method = PrincipalComponentMethod.Center; // PrincipalComponentMethod.Standardize
// Step 3. Compute the covariance matrix
// -------------------------------------
// Note: Accord.NET does not need to compute the covariance
// matrix in order to compute PCA. The framework uses the SVD
// method which is more numerically stable, but may require
// more processing or memory. In order to replicate the tutorial
// using covariance matrices, please see the next unit test.
// Create the analysis using the selected method
var pca = new KernelPrincipalComponentAnalysis(new Linear(), method);
// Compute it
pca.Learn(data);
// Step 4. Compute the eigenvectors and eigenvalues of the covariance matrix
// -------------------------------------------------------------------------
// Note: Since Accord.NET uses the SVD method rather than the Eigendecomposition
// method, the Eigenvalues are computed from the singular values. However, it is
// not the Eigenvalues themselves which are important, but rather their proportion:
// Those are the expected eigenvalues, in descending order:
double[] eigenvalues = { 1.28402771, 0.0490833989 };
// And this will be their proportion:
double[] proportion = eigenvalues.Divide(eigenvalues.Sum());
Assert.IsTrue(proportion.IsEqual(pca.ComponentProportions, rtol: 1e-9));
Assert.IsTrue(eigenvalues.IsEqual(pca.Eigenvalues.Divide(data.GetLength(0) - 1), rtol: 1e-5));
// Step 5. Deriving the new data set
// ---------------------------------
double[][] actual = pca.Transform(data);
// transformedData shown in pg. 18
double[,] expected = new double[, ]
{
{ 0.827970186, -0.175115307 },
{ -1.77758033, 0.142857227 },
{ 0.992197494, 0.384374989 },
{ 0.274210416, 0.130417207 },
{ 1.67580142, -0.209498461 },
{ 0.912949103, 0.175282444 },
{ -0.099109437, -0.349824698 },
{ -1.14457216, 0.046417258 },
{ -0.438046137, 0.017764629 },
{ -1.22382056, -0.162675287 },
}.Multiply(-1);
// Everything is correct (up to 8 decimal places)
Assert.IsTrue(expected.IsEqual(actual, atol: 1e-8));
// Finally, we can project all the data
double[][] output1 = pca.Transform(data);
// Or just its first components by setting
// NumberOfOutputs to the desired components:
pca.NumberOfOutputs = 1;
// And then calling transform again:
double[][] output2 = pca.Transform(data);
// We can also limit to 80% of explained variance:
pca.ExplainedVariance = 0.8;
// And then call transform again:
double[][] output3 = pca.Transform(data);
#endregion
actual = pca.Transform(data);
// transformedData shown in pg. 18
expected = new double[, ]
{
{ 0.827970186 },
{ -1.77758033, },
{ 0.992197494 },
{ 0.274210416 },
{ 1.67580142, },
{ 0.912949103 },
{ -0.099109437 },
{ -1.14457216, },
{ -0.438046137 },
{ -1.22382056, },
}.Multiply(-1);
// Everything is correct (up to 8 decimal places)
Assert.IsTrue(expected.IsEqual(actual, atol: 1e-8));
// Create the analysis using the selected method
pca = new KernelPrincipalComponentAnalysis()
{
Kernel = new Linear(),
Method = method,
NumberOfOutputs = 1
};
// Compute it
pca.Learn(data);
actual = pca.Transform(data);
// transformedData shown in pg. 18
expected = new double[, ]
{
{ 0.827970186 },
{ -1.77758033, },
{ 0.992197494 },
{ 0.274210416 },
{ 1.67580142, },
{ 0.912949103 },
{ -0.099109437 },
{ -1.14457216, },
{ -0.438046137 },
{ -1.22382056, },
}.Multiply(-1);
// Everything is correct (up to 8 decimal places)
Assert.IsTrue(expected.IsEqual(actual, atol: 1e-8));
}