// public override bool IsUncertain(params IKEPDist[] msgs){
// // Return true if at least one of the mappers is uncertain
// for(int i=0; i<onlineBayes.Length; i++){
// if(onlineBayes[i].IsUncertain(msgs)){
// return true;
// }
// }
// return false;
// }
private bool IsUncertain(Vector[] features)
{
// Return true if at least one of the mappers is uncertain
for (int i = 0; i < onlineBayes.Length; i++)
{
BayesLinRegFM bi = onlineBayes[i];
double[] thresh = bi.GetUncertaintyThreshold();
if (thresh.Length != 1)
{
throw new ArgumentException("Threshold from a Bayesian linear regression should have just one number.");
}
double t = thresh[0];
double[] un = bi.EstimateUncertainty();
if (un.Length != 1)
{
throw new ArgumentException("Uncertainty from a Bayesian linear regression should have just one number");
}
double u = un[0];
if (u >= t)
{
return(true);
}
}
return(false);
}
// return the expected number of incoming messages
// negative for any number.
// public abstract int NumInputMessages();
// Load a FeatureMap in Matlab represented by the input
// All FeatureMap objects can be serialized to struct with .toStruct()
public static VectorMapper FromMatlabStruct(MatlabStruct s)
{
string className = s.GetString("className");
VectorMapper map = null;
if (className.Equals("RandFourierGaussMVMap"))
{
map = RFGMVMap.FromMatlabStruct(s);
}
else if (className.Equals("CondFMFiniteOut"))
{
map = CondFMFiniteOut.FromMatlabStruct(s);
}
// else if(className.Equals("CondCholFiniteOut")){
// map = CondCholFiniteOut.FromMatlabStruct(s);
// }
else if (className.Equals(RFGJointKGG.MATLAB_CLASS))
{
map = RFGJointKGG.FromMatlabStruct(s);
}
else if (className.Equals(StackVectorMapper.MATLAB_CLASS))
{
map = StackVectorMapper.FromMatlabStruct(s);
}
else if (className.Equals(BayesLinRegFM.MATLAB_CLASS))
{
map = BayesLinRegFM.FromMatlabStruct(s);
}
else if (className.Equals(UAwareVectorMapper.MATLAB_CLASS))
{
map = UAwareVectorMapper.FromMatlabStruct(s);
}
else if (className.Equals(UAwareStackVectorMapper.MATLAB_CLASS))
{
map = UAwareStackVectorMapper.FromMatlabStruct(s);
}
else
{
throw new ArgumentException("Unknown className: " + className);
}
// else if(className.Equals("RFGSumEProdMap")){
//
// }else if(className.Equals("RFGEProdMap")){
//
// }else if(className.Equals("RFGJointEProdMap")){
//
// }else if(className.Equals("RFGProductEProdMap")){
//
// }
return(map);
}
public override void UpdateVectorMapper(Vector target, params IKEPDist[] msgs)
{
// update each internal mapper
if (target.Count != onlineBayes.Length)
{
throw new ArgumentException("Require target length == number of nested Bayes learners");
}
for (int i = 0; i < onlineBayes.Length; i++)
{
Vector oneDTarget = Vector.FromArray(new double[] { target[i] });
BayesLinRegFM map = onlineBayes[i];
map.UpdateVectorMapper(oneDTarget, msgs);
}
}
public static new BayesLinRegFM FromMatlabStruct(MatlabStruct s)
{
// s.className=class(this);
// s.featureMap=this.featureMap.toStruct();
// %s.regParam=this.regParam;
// s.mapMatrix=this.mapMatrix;
// s.posteriorCov = this.posteriorCov;
// s.noise_var = this.noise_var;
string className = s.GetString("className");
if (!className.Equals(MATLAB_CLASS))
{
throw new ArgumentException("The input does not represent a " + MATLAB_CLASS);
}
MatlabStruct fmStruct = s.GetStruct("featureMap");
RandomFeatureMap featureMap = RandomFeatureMap.FromMatlabStruct(fmStruct);
// This is the same as a posterior mean
Vector mapMatrix = s.Get1DVector("mapMatrix");
if (mapMatrix.Count != featureMap.GetOutputDimension())
{
throw new ArgumentException("mapMatrix and featureMap's dimenions are incompatible.");
}
Matrix postCov = s.GetMatrix("posteriorCov");
if (postCov.Cols != featureMap.GetOutputDimension())
{
throw new ArgumentException("posterior covariance and featureMap's dimenions are incompatible.");
}
double noise_var = s.GetDouble("noise_var");
Vector crossCorr = s.Get1DVector("crossCorrelation");
var bayes = new BayesLinRegFM();
bayes.featureMap = featureMap;
bayes.posteriorMean = mapMatrix;
bayes.posteriorCov = postCov;
bayes.noiseVar = noise_var;
bayes.crossCorr = crossCorr;
// No need to do the initial batch train because we loaded the result
// from .mat.
bayes.WillNeedInitialTrain = false;
return(bayes);
}
