/// <summary>
/// Reads a problem from a stream.
/// </summary>
/// <param name="stream">Stream to read from</param>
/// <returns>The problem</returns>
public static Problem Read(Stream stream)
{
TemporaryCulture.Start();
StreamReader input = new StreamReader(stream);
List <double> vy = new List <double>();
List <Node[]> vx = new List <Node[]>();
int max_index = 0;
while (input.Peek() > -1)
{
string[] parts = input.ReadLine().Trim().Split();
vy.Add(double.Parse(parts[0]));
int m = parts.Length - 1;
Node[] x = new Node[m];
for (int j = 0; j < m; j++)
{
x[j] = new Node();
string[] nodeParts = parts[j + 1].Split(':');
x[j].Index = int.Parse(nodeParts[0]);
x[j].Value = double.Parse(nodeParts[1]);
}
if (m > 0)
{
max_index = Math.Max(max_index, x[m - 1].Index);
}
vx.Add(x);
}
TemporaryCulture.Stop();
return(new Problem(vy.Count, vy.ToArray(), vx.ToArray(), max_index));
}
/// <summary>
/// Reads a Range transform from a stream.
/// </summary>
/// <param name="stream">The stream to read from</param>
/// <returns>The Range transform</returns>
public static RangeTransform Read(Stream stream)
{
TemporaryCulture.Start();
StreamReader input = new StreamReader(stream);
int length = int.Parse(input.ReadLine());
double[] inputStart = new double[length];
double[] inputScale = new double[length];
string[] parts = input.ReadLine().Split();
for (int i = 0; i < length; i++)
{
inputStart[i] = double.Parse(parts[i]);
}
parts = input.ReadLine().Split();
for (int i = 0; i < length; i++)
{
inputScale[i] = double.Parse(parts[i]);
}
parts = input.ReadLine().Split();
double outputStart = double.Parse(parts[0]);
double outputScale = double.Parse(parts[1]);
TemporaryCulture.Stop();
return(new RangeTransform(inputStart, inputScale, outputStart, outputScale, length));
}
/// <summary>
/// Saves the transform to the disk. The samples are not stored, only the
/// statistics.
/// </summary>
/// <param name="stream">The destination stream</param>
/// <param name="transform">The transform</param>
public static void Write(Stream stream, GaussianTransform transform)
{
TemporaryCulture.Start();
StreamWriter output = new StreamWriter(stream);
output.WriteLine(transform._means.Length);
for (int i = 0; i < transform._means.Length; i++)
{
output.WriteLine("{0} {1}", transform._means[i], transform._stddevs[i]);
}
output.Flush();
TemporaryCulture.Stop();
}
/// <summary>
/// Writes a problem to a stream.
/// </summary>
/// <param name="stream">The stream to write the problem to.</param>
/// <param name="problem">The problem to write.</param>
public static void Write(Stream stream, Problem problem)
{
TemporaryCulture.Start();
StreamWriter output = new StreamWriter(stream);
for (int i = 0; i < problem.Count; i++)
{
output.Write(problem.Y[i]);
for (int j = 0; j < problem.X[i].Length; j++)
{
output.Write(" {0}:{1}", problem.X[i][j].Index, problem.X[i][j].Value);
}
output.WriteLine();
}
output.Flush();
TemporaryCulture.Stop();
}
/// <summary>
/// Reads a GaussianTransform from the provided stream.
/// </summary>
/// <param name="stream">The source stream</param>
/// <returns>The transform</returns>
public static GaussianTransform Read(Stream stream)
{
TemporaryCulture.Start();
StreamReader input = new StreamReader(stream);
int length = int.Parse(input.ReadLine(), CultureInfo.InvariantCulture);
double[] means = new double[length];
double[] stddevs = new double[length];
for (int i = 0; i < length; i++)
{
string[] parts = input.ReadLine().Split();
means[i] = double.Parse(parts[0], CultureInfo.InvariantCulture);
stddevs[i] = double.Parse(parts[1], CultureInfo.InvariantCulture);
}
TemporaryCulture.Stop();
return(new GaussianTransform(means, stddevs));
}
/// <summary>
/// Writes this Range transform to a stream.
/// </summary>
/// <param name="stream">The stream to write to</param>
/// <param name="r">The range to write</param>
public static void Write(Stream stream, RangeTransform r)
{
TemporaryCulture.Start();
StreamWriter output = new StreamWriter(stream);
output.WriteLine(r._length);
output.Write(r._inputStart[0]);
for (int i = 1; i < r._inputStart.Length; i++)
{
output.Write(" " + r._inputStart[i]);
}
output.WriteLine();
output.Write(r._inputScale[0]);
for (int i = 1; i < r._inputScale.Length; i++)
{
output.Write(" " + r._inputScale[i]);
}
output.WriteLine();
output.WriteLine("{0} {1}", r._outputStart, r._outputScale);
output.Flush();
TemporaryCulture.Stop();
}
/// <summary>
/// Writes a model to the provided stream.
/// </summary>
/// <param name="stream">The output stream</param>
/// <param name="model">The model to write</param>
public static void Write(Stream stream, Model model)
{
TemporaryCulture.Start();
StreamWriter output = new StreamWriter(stream);
Parameter param = model.Parameter;
output.Write("svm_type " + param.SvmType + "\n");
output.Write("kernel_type " + param.KernelType + "\n");
if (param.KernelType == KernelType.POLY)
{
output.Write("degree " + param.Degree + "\n");
}
if (param.KernelType == KernelType.POLY || param.KernelType == KernelType.RBF || param.KernelType == KernelType.SIGMOID)
{
output.Write("gamma " + param.Gamma + "\n");
}
if (param.KernelType == KernelType.POLY || param.KernelType == KernelType.SIGMOID)
{
output.Write("coef0 " + param.Coefficient0 + "\n");
}
int nr_class = model.NumberOfClasses;
int l = model.SupportVectorCount;
output.Write("nr_class " + nr_class + "\n");
output.Write("total_sv " + l + "\n");
{
output.Write("rho");
for (int i = 0; i < nr_class * (nr_class - 1) / 2; i++)
{
output.Write(" " + model.Rho[i]);
}
output.Write("\n");
}
if (model.ClassLabels != null)
{
output.Write("label");
for (int i = 0; i < nr_class; i++)
{
output.Write(" " + model.ClassLabels[i]);
}
output.Write("\n");
}
if (model.PairwiseProbabilityA != null)
// regression has probA only
{
output.Write("probA");
for (int i = 0; i < nr_class * (nr_class - 1) / 2; i++)
{
output.Write(" " + model.PairwiseProbabilityA[i]);
}
output.Write("\n");
}
if (model.PairwiseProbabilityB != null)
{
output.Write("probB");
for (int i = 0; i < nr_class * (nr_class - 1) / 2; i++)
{
output.Write(" " + model.PairwiseProbabilityB[i]);
}
output.Write("\n");
}
if (model.NumberOfSVPerClass != null)
{
output.Write("nr_sv");
for (int i = 0; i < nr_class; i++)
{
output.Write(" " + model.NumberOfSVPerClass[i]);
}
output.Write("\n");
}
output.Write("SV\n");
double[][] sv_coef = model.SupportVectorCoefficients;
Node[][] SV = model.SupportVectors;
for (int i = 0; i < l; i++)
{
for (int j = 0; j < nr_class - 1; j++)
{
output.Write(sv_coef[j][i] + " ");
}
Node[] p = SV[i];
if (p.Length == 0)
{
output.WriteLine();
continue;
}
if (param.KernelType == KernelType.PRECOMPUTED)
{
output.Write("0:{0}", (int)p[0].Value);
}
else
{
output.Write("{0}:{1}", p[0].Index, p[0].Value);
for (int j = 1; j < p.Length; j++)
{
output.Write(" {0}:{1}", p[j].Index, p[j].Value);
}
}
output.WriteLine();
}
output.Flush();
TemporaryCulture.Stop();
}
/// <summary>
/// Reads a Model from the provided stream.
/// </summary>
/// <param name="stream">The stream from which to read the Model.</param>
/// <returns>the Model</returns>
public static Model Read(Stream stream)
{
TemporaryCulture.Start();
StreamReader input = new StreamReader(stream);
// read parameters
Model model = new Model();
Parameter param = new Parameter();
model.Parameter = param;
model.Rho = null;
model.PairwiseProbabilityA = null;
model.PairwiseProbabilityB = null;
model.ClassLabels = null;
model.NumberOfSVPerClass = null;
bool headerFinished = false;
while (!headerFinished)
{
string line = input.ReadLine();
string cmd, arg;
int splitIndex = line.IndexOf(' ');
if (splitIndex >= 0)
{
cmd = line.Substring(0, splitIndex);
arg = line.Substring(splitIndex + 1);
}
else
{
cmd = line;
arg = "";
}
arg = arg.ToLower();
int i, n;
switch (cmd)
{
case "svm_type":
param.SvmType = (SvmType)Enum.Parse(typeof(SvmType), arg.ToUpper());
break;
case "kernel_type":
param.KernelType = (KernelType)Enum.Parse(typeof(KernelType), arg.ToUpper());
break;
case "degree":
param.Degree = int.Parse(arg);
break;
case "gamma":
param.Gamma = double.Parse(arg);
break;
case "coef0":
param.Coefficient0 = double.Parse(arg);
break;
case "nr_class":
model.NumberOfClasses = int.Parse(arg);
break;
case "total_sv":
model.SupportVectorCount = int.Parse(arg);
break;
case "rho":
n = model.NumberOfClasses * (model.NumberOfClasses - 1) / 2;
model.Rho = new double[n];
string[] rhoParts = arg.Split();
for (i = 0; i < n; i++)
{
model.Rho[i] = double.Parse(rhoParts[i]);
}
break;
case "label":
n = model.NumberOfClasses;
model.ClassLabels = new int[n];
string[] labelParts = arg.Split();
for (i = 0; i < n; i++)
{
model.ClassLabels[i] = int.Parse(labelParts[i]);
}
break;
case "probA":
n = model.NumberOfClasses * (model.NumberOfClasses - 1) / 2;
model.PairwiseProbabilityA = new double[n];
string[] probAParts = arg.Split();
for (i = 0; i < n; i++)
{
model.PairwiseProbabilityA[i] = double.Parse(probAParts[i]);
}
break;
case "probB":
n = model.NumberOfClasses * (model.NumberOfClasses - 1) / 2;
model.PairwiseProbabilityB = new double[n];
string[] probBParts = arg.Split();
for (i = 0; i < n; i++)
{
model.PairwiseProbabilityB[i] = double.Parse(probBParts[i]);
}
break;
case "nr_sv":
n = model.NumberOfClasses;
model.NumberOfSVPerClass = new int[n];
string[] nrsvParts = arg.Split();
for (i = 0; i < n; i++)
{
model.NumberOfSVPerClass[i] = int.Parse(nrsvParts[i]);
}
break;
case "SV":
headerFinished = true;
break;
default:
throw new Exception("Unknown text in model file");
}
}
// read sv_coef and SV
int m = model.NumberOfClasses - 1;
int l = model.SupportVectorCount;
model.SupportVectorCoefficients = new double[m][];
for (int i = 0; i < m; i++)
{
model.SupportVectorCoefficients[i] = new double[l];
}
model.SupportVectors = new Node[l][];
for (int i = 0; i < l; i++)
{
string[] parts = input.ReadLine().Trim().Split();
for (int k = 0; k < m; k++)
{
model.SupportVectorCoefficients[k][i] = double.Parse(parts[k]);
}
int n = parts.Length - m;
model.SupportVectors[i] = new Node[n];
for (int j = 0; j < n; j++)
{
string[] nodeParts = parts[m + j].Split(':');
model.SupportVectors[i][j] = new Node();
model.SupportVectors[i][j].Index = int.Parse(nodeParts[0]);
model.SupportVectors[i][j].Value = double.Parse(nodeParts[1]);
}
}
TemporaryCulture.Stop();
return(model);
}
