public LogPrior(LogPrior.LogPriorType type, double sigma, double epsilon)
{
this.type = type;
if (type != LogPrior.LogPriorType.Adapt)
{
SetSigma(sigma);
SetEpsilon(epsilon);
}
}
/// <summary>
/// IMPORTANT NOTE: This constructor allows non-uniform regularization, but it
/// transforms the inputs C (like the machine learning people like) to sigma
/// (like we NLP folks like).
/// </summary>
/// <remarks>
/// IMPORTANT NOTE: This constructor allows non-uniform regularization, but it
/// transforms the inputs C (like the machine learning people like) to sigma
/// (like we NLP folks like). C = 1/\sigma^2
/// </remarks>
public LogPrior(double[] C)
{
// this is the C variable in CSFoo's MM paper C = 1/\sigma^2
// private double[] regularizationHyperparameters = null;
// public double[] getRegularizationHyperparameters() {
// return regularizationHyperparameters;
// }
//
// public void setRegularizationHyperparameters(
// double[] regularizationHyperparameters) {
// this.regularizationHyperparameters = regularizationHyperparameters;
// }
this.type = LogPrior.LogPriorType.MultipleQuadratic;
double[] sigmaSqM = new double[C.Length];
for (int i = 0; i < C.Length; i++)
{
sigmaSqM[i] = 1.0 / C[i];
}
this.sigmaSqM = sigmaSqM;
SetSigmaSquaredM(sigmaSqM);
}
public LogPrior(LogPrior.LogPriorType type)
: this(type, 1.0, 0.1)
{
}
