public TablRow(DaoRec rec)
{
foreach (Field f in rec.Recordset.Fields)
{
Means.Add(f.Name, rec.GetMean(f.Type.ToDataType(), f.Name));
}
}
/// <summary>
/// Generates new normal distribution settings as a sum of correlated normal distributions with coefficients <paramref name="coeffs"/> (c1*A+c2*B+c3*C etc.)
/// </summary>
/// <param name="coeffs">Coefficients vector that will be applied to the sum of normal distributions.</param>
/// <returns>Normal distribution settings instance.</returns>
public NormalDistributionSettings GetUnivariateDistributionSettings(double[] coeffs)
{
if (coeffs == null)
{
coeffs = Vector.Ones(Dimension);
}
if (coeffs.Length != Dimension)
{
throw new DistributionsArgumentException(DistributionsArgumentExceptionType.VectorOfCoeffitientsMustBeEqualToDimension);
}
if (coeffs.All(x => x == 1))
{
return(new NormalDistributionSettings(Means.Sum(), Math.Sqrt(CovarianceMatrix.Sum())));
}
double[,] ltf = Chol.LeftTriangularFactor;
var weighted = Matrix.Diagonal(coeffs).Dot(ltf);
var colSumPow = weighted.Transpose().Dot(Vector.Ones(Dimension)).Pow(2);
double variance = colSumPow.Sum();
double mean = Elementwise.Multiply(coeffs, Means).Sum();
return(new NormalDistributionSettings(mean, Math.Sqrt(variance)));
}
private Pipeline CreateClassifier()
{
if (NumberOfOutputs == 0)
{
return(null);
}
return(new Pipeline()
{
NumberOfInputs = NumberOfInputs,
NumberOfOutputs = NumberOfClasses,
First = new MultivariateKernelRegression()
{
Weights = DiscriminantVectors,
Intercept = Means.DotWithTransposed(DiscriminantVectors).Multiply(-1),
BasisVectors = input,
Kernel = Kernel,
NumberOfInputs = NumberOfInputs,
NumberOfOutputs = NumberOfOutputs,
},
Second = new MinimumMeanDistanceClassifier()
{
Means = projectedMeans,
NumberOfInputs = NumberOfOutputs,
NumberOfOutputs = NumberOfClasses,
}
});
}
private MixtureModel(IList <int> k, IList <int> n, double[] startingMeans) :
this(k, n, startingMeans, new double[startingMeans.Length])
{
int minIndex = Array.IndexOf(Means, Means.Min());
MixtureWeights[minIndex] = 0.99;
for (int i = 0; i < MixtureWeights.Length; i++)
{
if (i == minIndex)
{
continue;
}
MixtureWeights[i] = 0.01 / (MixtureWeights.Length - 1);
}
}
public void Predict(TimeSpan ts)
{
lastTimeSpan = ts;
var lastdate = Means.Last().Key;
var newdate = lastdate + ts;
var xp = filter.Predict(x, P, f, Q, lastTimeSpan.Ticks);
var x_ = x.ToArray();
Means[newdate] = x_;
var P_ = P.ToArray();
CoVariances[newdate] = P_;
}
public void Predict(out double[] x_, out double[,] P_, TimeSpan ts)
{
lastTimeSpan = ts;
var lastdate = Means.Last().Key;
var newdate = lastdate + ts;
Q = q.Matrix(ts.Ticks);
var xp = FilterSharp.Predict(x, P, f, Q, lastTimeSpan.Ticks);
x_ = x.ToArray();
Means[newdate] = x_;
P_ = P.ToArray();
CoVariances[newdate] = P_;
}
public Tuple <DateTime, double>[] PositionMeans()
{
return(Means.Select(_ => new Tuple <DateTime, double>(_.Key, _.Value[0])).ToArray());
}
