// /// <summary>
// /// Gets the value of information.
// /// </summary>
// /// <value>The value of information.</value>
// public double[] GetValueOfInformation()
// {
// double jall = JAll();
// return Unlabelled.Select(index => VOI(jall, index)).ToArray();
// }
/// <summary>
/// Gets the argument maxising the Value of information.
/// </summary>
/// <param name="activityPosteriors">Activity posteriors.</param>
/// <param name="priors">Priors.</param>
/// <param name="argMax">Argument max.</param>
/// <param name="maxVal">Max value.</param>
public override void GetArgMaxVOI(Bernoulli[] activityPosteriors, Marginals priors, out int argMax, out double maxVal)
{
hypothesisActivityPosteriors = activityPosteriors.Select(ia => new Bernoulli(ia)).ToArray();
double jall = JAll();
var unlabelled = Unlabelled.ToArray();
argMax = -1;
maxVal = Reversed ? double.PositiveInfinity : double.NegativeInfinity;
var signs = new double[unlabelled.Length];
var vois = new double[unlabelled.Length];
for (int i = 0; i < unlabelled.Length; i++)
{
var index = unlabelled[i];
vois[i] = VOI(jall, index, activityPosteriors, priors);
signs[i] = Math.Sign(vois[i]) / 2.0 + 0.5;
//Console.Write( "." );
//Console.WriteLine( "y_true: {0}", labels[0][ind] );
//Console.WriteLine( "y_hat: {0}", probs[ind] );
//Console.WriteLine( "VOJ_{0}: {1}", ind, voi );
//Console.WriteLine();
if (Reversed)
{
if (vois[i] < maxVal || argMax < 0)
{
maxVal = vois[i];
argMax = index;
}
}
else
{
if (vois[i] > maxVal || argMax < 0)
{
maxVal = vois[i];
argMax = index;
}
}
}
//Console.WriteLine();
//Console.WriteLine( "\n+ivity: {0}", signs.Average() );
}
private double JAll_j(int index, Bernoulli[] activityPosteriors, Marginals priors)
{
// var prevProbs = activityPosteriors.Select(ia => new Bernoulli(ia)).ToArray();
hypothesisActivityPosteriors = activityPosteriors.Select(ia => new Bernoulli(ia)).ToArray();
// Get datum
// var datum = DataSet.GetSubSet(0, index);
bool trueLabel = DataSet.Labels[0][index];
// Create copies of the Labelled an Unlabelled sets
var labelled = new HashSet<int>(Labelled);
var unlabelled = new HashSet<int>(Unlabelled);
labelled.Add(index);
unlabelled.Remove(index);
// datum.Labels[0][0] = true;
DataSet.Labels[0][index] = true;
// Learn as if positive
// var positivePosteriors = TrainModel.Train(datum, priors, 1);
Marginals positivePosteriors = priors;
try
{
positivePosteriors = TrainModel.Train(DataSet.GetSubSet(0, index), priors, 1);
}
catch (ImproperMessageException)
{
// As fallback use priors
}
// recompute probabilities
CalculateProbabilities(positivePosteriors);
//var jjposl = JL(labelled);
//var jjposu = JU(unlabelled);
var Jjpos = (JAll()) * (1.0 - hypothesisActivityPosteriors[index].GetMean());
// Restore posteriors
labelled.Add(index);
unlabelled.Remove(index);
// datum.Labels[0][0] = false;
DataSet.Labels[0][index] = false;
// Learn as if negative
// var negativePosteriors = TrainModel.Train(datum, priors, 1);
Marginals negativePosteriors = priors;
try
{
negativePosteriors = TrainModel.Train(DataSet.GetSubSet(0, index), priors, 1);
}
catch (ImproperMessageException)
{
// As fallback use priors
}
// recompute probabilities
CalculateProbabilities(negativePosteriors);
//var jjnegl = JL(labelled);
//var jjnegu = JU(unlabelled);
var Jjneg = (JAll()) * (hypothesisActivityPosteriors[index].GetMean());
// restore posteriors
// activityPosteriors = prevProbs;
DataSet.Labels[0][index] = trueLabel;
var voi = Jjpos + Jjneg;
return voi;
}
/// <summary>
/// Tests the binary hidden markov model.
/// </summary>
public static void TestBinaryHiddenMarkovModel()
{
// fix random seed
Rand.Restart(12347);
// model size
const int T = 100;
const int K = 2;
// set hyperparameters
var probInitPriorObs = Dirichlet.Uniform(K);
var cptTransPriorObs = Enumerable.Repeat(Dirichlet.Uniform(K), K).ToArray();
var emitPriorObs = Enumerable.Repeat(new Beta(1, 1), K).ToArray();
// sample model parameters
var init = probInitPriorObs.Sample().ToArray();
var trans = new double[K][];
for (int i = 0; i < K; i++)
{
trans[i] = cptTransPriorObs[i].Sample().ToArray();
}
var emit = new double[K];
for (int i = 0; i < K; i++)
{
emit[i] = emitPriorObs[i].Sample();
}
// print parameters
var modelForPrinting = new BinaryHMM();
modelForPrinting.CreateModel(T, K, 1, false);
modelForPrinting.SetParameters(init, trans, emit, null);
Console.WriteLine("parameters:");
modelForPrinting.PrintParameters();
Console.WriteLine();
// create distributions for sampling
var initDist = new Discrete(init);
var transDist = new Discrete[K];
for (int i = 0; i < K; i++)
{
transDist[i] = new Discrete(trans[i]);
}
var emitDist = new Bernoulli[K];
for (int i = 0; i < K; i++)
{
emitDist[i] = new Bernoulli(emit[i]);
}
// calculate order of actual states
var actualStateOrder = emitDist.Select((ia, i) => new { ia, i }).OrderBy(pair => pair.ia.GetMean()).Select(pair => pair.i).ToArray();
Console.WriteLine("actualStateOrder");
Console.WriteLine(string.Join(",", actualStateOrder));
Console.WriteLine();
// sample state and emission data
var actualStates = new int[T];
var emissions = new bool[T];
actualStates[0] = initDist.Sample();
emissions[0] = emitDist[actualStates[0]].Sample();
for (int i = 1; i < T; i++)
{
actualStates[i] = transDist[actualStates[i - 1]].Sample();
emissions[i] = emitDist[actualStates[i]].Sample();
}
Console.WriteLine("sample data:");
Console.WriteLine(string.Join(",", actualStates));
Console.WriteLine();
// infer model parameters, states and model evidence given priors and emission data
var model = new BinaryHMM();
model.CreateModel(T, K, 1, false);
model.SetPriors(probInitPriorObs, cptTransPriorObs, emitPriorObs, null);
model.ObserveData(emissions);
model.InitialiseStatesRandomly();
model.InferPosteriors();
Console.WriteLine("model likelihood: " + model.modelEvidencePosterior);
var mapStatesDistr = model.statesPosterior;
var mapStates = mapStatesDistr.Select(s => s.GetMode()).ToArray();
Console.WriteLine();
// print maximum a priori states
Console.WriteLine("statesMAP");
Console.WriteLine(string.Join(",", mapStates));
Console.WriteLine();
// print posterior distributions
Console.WriteLine("posteriors");
model.PrintPosteriors();
Console.WriteLine();
// calculate order of MAP states
var mapStateOrder = ArgSort <Beta, double>(model.emitMeanPosterior);
Console.WriteLine(string.Join(",", mapStateOrder));
Console.WriteLine();
// accuracy of MAP estimates
int correctStates = actualStates.Where((t, i) => actualStateOrder[t] == mapStateOrder[mapStates[i]]).Count();
Console.WriteLine("correctStates: " + correctStates + " / " + actualStates.Length);
Console.WriteLine();
Console.WriteLine("------------------\n");
}