public void ReadSampleTest()
{
// http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#iris
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Accord.Tests.IO.Properties.Resources.iris_scale));
// Suppose we are going to read a sparse sample file containing
// samples which have an actual dimension of 4. Since the samples
// are in a sparse format, each entry in the file will probably
// have a much lesser number of elements.
int sampleSize = 4;
// Create a new Sparse Sample Reader to read any given file,
// passing the correct dense sample size in the constructor
SparseReader reader = new SparseReader(file, Encoding.Default, sampleSize);
// Declare some variables to receive each current sample
int label = 0;
string description;
double[] sample;
// Read a sample from the file
var r = reader.ReadDense();
sample = r.Item1;
label = (int)r.Item2;
description = reader.SampleDescriptions[0];
Assert.AreEqual(1, label);
Assert.AreEqual(String.Empty, description);
Assert.AreEqual(4, sample.Length);
Assert.AreEqual(-0.555556, sample[0], 0.0001);
Assert.AreEqual(+0.250000, sample[1], 0.0001);
Assert.AreEqual(-0.864407, sample[2], 0.0001);
Assert.AreEqual(-0.916667, sample[3], 0.0001);
var s = reader.ReadSparse();
sample = s.Item1.ToSparse();
label = (int)s.Item2;
description = reader.SampleDescriptions[0];
Assert.AreEqual(1, label);
Assert.AreEqual(String.Empty, description);
Assert.AreEqual(8, sample.Length);
Assert.AreEqual(0, sample[0], 0.0001);
Assert.AreEqual(-0.666667, sample[1], 0.0001);
Assert.AreEqual(1, sample[2], 0.0001);
Assert.AreEqual(-0.166667, sample[3], 0.0001);
Assert.AreEqual(2, sample[4], 0.0001);
Assert.AreEqual(-0.864407, sample[5], 0.0001);
Assert.AreEqual(3, sample[6], 0.0001);
Assert.AreEqual(-0.916667, sample[7], 0.0001);
int count = 2;
// Read all samples from the file
while (!reader.EndOfStream)
{
reader.SampleDescriptions.Clear();
r = reader.ReadDense();
sample = r.Item1;
label = (int)r.Item2;
description = reader.SampleDescriptions[0];
Assert.IsTrue(label >= 0 && label <= 3);
Assert.IsTrue(description == String.Empty);
Assert.AreEqual(4, sample.Length);
count++;
}
Assert.AreEqual(150, count);
}
public void GuessNumberOfDimensionsTest()
{
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.iris_scale));
SparseReader reader = new SparseReader(file, Encoding.Default);
Assert.AreEqual(4, reader.Dimensions);
var r = reader.ReadDenseToEnd();
double[][] samples = r.Item1;
int[] labels = r.Item2.ToInt32();
string[] descriptions = reader.SampleDescriptions.ToArray();
Assert.AreEqual(150, samples.Length);
for (int i = 0; i < 150; i++)
{
Assert.IsTrue(labels[i] >= 0 && labels[i] <= 3);
Assert.IsTrue(descriptions[i] == String.Empty);
Assert.AreEqual(4, samples[i].Length);
}
}
public void GuessDimensionsInMiddleRunTest()
{
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.a9a_train));
SparseReader reader = new SparseReader(file, Encoding.Default);
var r = reader.ReadDenseToEnd();
double[][] samples = r.Item1;
int[] labels = r.Item2.ToInt32();
string[] descriptions = reader.SampleDescriptions.ToArray();
Assert.AreEqual(26049, samples.Length);
for (int i = 0; i < labels.Length; i++)
{
Assert.IsTrue(labels[i] == -1 || labels[i] == 1);
Assert.IsTrue(descriptions[i] == String.Empty);
Assert.AreEqual(123, samples[i].Length);
}
}
public void DimensionsTest()
{
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.a9a_train));
SparseReader reader = new SparseReader(file, Encoding.Default);
Assert.AreEqual(123, reader.Dimensions);
int[] labels = null;
string[] descriptions = null;
double[][] samples = reader.ReadToEnd(out labels, out descriptions);
Assert.AreEqual(26049, samples.Length);
for (int i = 0; i < labels.Length; i++)
{
Assert.IsTrue(labels[i] == -1 || labels[i] == 1);
Assert.IsTrue(descriptions[i] == String.Empty);
Assert.AreEqual(123, samples[i].Length);
}
}
public void ReadAllTest()
{
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.iris_scale));
// Suppose we are going to read a sparse sample file containing
// samples which have an actual dimension of 4. Since the samples
// are in a sparse format, each entry in the file will probably
// have a much lesser number of elements.
int sampleSize = 4;
// Create a new Sparse Sample Reader to read any given file,
// passing the correct dense sample size in the constructor
SparseReader reader = new SparseReader(file, Encoding.Default, sampleSize);
// Declare a vector to obtain the label
// of each of the samples in the file
// Declare a vector to obtain the description (or comments)
// about each of the samples in the file, if present.
// Read the sparse samples and store them in a dense vector array
var r = reader.ReadDenseToEnd();
double[][] samples = r.Item1;
int[] labels = r.Item2.ToInt32();
string[] descriptions = reader.SampleDescriptions.ToArray();
Assert.AreEqual(150, samples.Length);
for (int i = 0; i < 150; i++)
{
Assert.IsTrue(labels[i] >= 0 && labels[i] <= 3);
Assert.IsTrue(descriptions[i] == String.Empty);
Assert.AreEqual(4, samples[i].Length);
}
}
/// <summary>
/// Reads a problem specified in LibSVM's sparse format.
/// </summary>
///
public static Problem read_problem(string filename, double bias)
{
// Create a LibSVM's sparse data reader
var reader = new SparseReader(filename);
if (bias > 0)
reader.Intercept = bias;
double[] y; // read inputs and labels
double[][] x = reader.ReadToEnd(out y);
return new Problem()
{
Dimensions = reader.Dimensions,
Inputs = x,
Outputs = y,
};
}
public void SparseLinearTest()
{
MulticlassSupportVectorMachine<Linear> svm1;
MulticlassSupportVectorMachine<Linear, Sparse<double>> svm2;
{
Accord.Math.Random.Generator.Seed = 0;
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.iris_scale));
var reader = new SparseReader(file, Encoding.Default);
var samples = reader.ReadDenseToEnd();
double[][] x = samples.Item1;
int[] y = samples.Item2.ToMulticlass();
var learner = new MulticlassSupportVectorLearning<Linear>()
{
Learner = (p) => new LinearDualCoordinateDescent<Linear>()
};
svm1 = learner.Learn(x, y);
}
{
Accord.Math.Random.Generator.Seed = 0;
MemoryStream file = new MemoryStream(
Encoding.Default.GetBytes(Resources.iris_scale));
// Create a new Sparse Sample Reader to read any given file,
// passing the correct dense sample size in the constructor
var reader = new SparseReader(file, Encoding.Default);
var samples = reader.ReadSparseToEnd();
Sparse<double>[] x = samples.Item1;
int[] y = samples.Item2.ToMulticlass();
var learner = new MulticlassSupportVectorLearning<Linear, Sparse<double>>()
{
Learner = (p) => new LinearDualCoordinateDescent<Linear, Sparse<double>>()
};
svm2 = learner.Learn(x, y);
}
Assert.AreEqual(svm1.Models.Length, svm2.Models.Length);
for (int i = 0; i < svm1.Models.Length; i++)
{
var ma = svm1[i].Value;
var mb = svm2[i].Value;
Assert.IsTrue(ma.Weights.IsEqual(mb.Weights));
Assert.AreEqual(ma.SupportVectors.Length, mb.SupportVectors.Length);
for (int j = 0; j < ma.SupportVectors.Length; j++)
{
double[] expected = ma.SupportVectors[j];
double[] actual = mb.SupportVectors[j].ToDense(4);
Assert.IsTrue(expected.IsEqual(actual, 1e-5));
}
}
}
public static DataTable Load(Stream stream, TableFormat format)
{
switch (format)
{
case TableFormat.SerializedXml:
{
XmlSerializer serializer = new XmlSerializer(typeof(DataTable));
return (DataTable)serializer.Deserialize(stream);
}
case TableFormat.SerializedBin:
{
BinaryFormatter serializer = new BinaryFormatter();
return (DataTable)serializer.Deserialize(stream);
}
case TableFormat.OctaveMatFile:
{
MatReader reader = new MatReader(stream);
return reader.Fields.First().Value.GetValue<double[,]>().ToTable();
}
case TableFormat.OpenDocument:
{
ExcelReader reader = new ExcelReader(stream, true);
string ws = reader.GetWorksheetList().First();
return reader.GetWorksheet(ws);
}
case TableFormat.OlderExcel:
{
ExcelReader reader = new ExcelReader(stream, false);
string ws = reader.GetWorksheetList().First();
return reader.GetWorksheet(ws);
}
case TableFormat.Csv:
{
CsvReader reader = new CsvReader(stream, true);
return reader.ToTable();
}
case TableFormat.Tsv:
{
CsvReader reader = new CsvReader(stream, true, '\t');
return reader.ToTable();
}
case TableFormat.LibSVM:
{
SparseReader reader = new SparseReader(stream);
return reader.ToTable();
}
case TableFormat.Idx:
{
IdxReader reader = new IdxReader(stream);
return reader.ReadToEndAsVectors().ToTable();
}
case TableFormat.CSharp:
throw new NotSupportedException();
}
}
/// <summary>
/// Reads a problem specified in LibSVM's sparse format.
/// </summary>
///
public static Problem read_problem(string filename, double bias)
{
// Create a LibSVM's sparse data reader
var reader = new SparseReader(filename);
if (bias > 0)
reader.Intercept = bias;
var r = reader.ReadDenseToEnd();
double[][] x = r.Item1;
double[] y = r.Item2;
return new Problem()
{
Dimensions = reader.Dimensions,
Inputs = x,
Outputs = y,
};
}
private static void cancer()
{
// Create a new LibSVM sparse format data reader
// to read the Wisconsin's Breast Cancer dataset
//
var reader = new SparseReader("examples-sparse.txt");
int[] outputs; // Read the classification problem into dense memory
double[][] inputs = reader.ReadToEnd(sparse: false, labels: out outputs);
// The dataset has output labels as 4 and 2. We have to convert them
// into negative and positive labels so they can be properly processed.
//
outputs = outputs.Apply(x => x == 2 ? -1 : +1);
// Create a new linear-SVM for the problem dimensions
var svm = new SupportVectorMachine(inputs: reader.Dimensions);
// Create a learning algorithm for the problem's dimensions
var teacher = new LinearDualCoordinateDescent(svm, inputs, outputs)
{
Loss = Loss.L2,
Complexity = 1000,
Tolerance = 1e-5
};
// Learn the classification
double error = teacher.Run();
// Compute the machine's answers for the learned inputs
int[] answers = inputs.Apply(x => Math.Sign(svm.Compute(x)));
// Create a confusion matrix to show the machine's performance
var m = new ConfusionMatrix(predicted: answers, expected: outputs);
// Show it onscreen
DataGridBox.Show(new ConfusionMatrixView(m));
}
