/// <summary>
/// VMP Messages to Uses
/// </summary>
/// <param name="Uses">Nonconjugate messages from uses</param>
/// <param name="Def">Gaussian message from Defintion</param>
/// <param name="result">Previous message</param>
/// <returns></returns>
public static DistributionStructArray<Gaussian,double> UsesAverageLogarithm(DistributionStructArray<NonconjugateGaussian,double>[] Uses, DistributionStructArray<Gaussian,double> Def, DistributionStructArray<Gaussian,double> result)
{
// multiply the messages together
DistributionStructArray<NonconjugateGaussian,double> prod = Uses[0];
for (int i = 1; i < Uses.Length; i++)
prod.SetToProduct(prod, Uses[i]);
// convert to Gaussians
for (int i = 0; i < prod.Count; i++)
{
result[i] = prod[i].GetGaussian(true);
if ((!result[i].IsProper() && !result[i].IsUniform()) || double.IsNaN(result[i].Precision))
throw new ApplicationException("UsesAverageLogarithm ");
}
result.SetToProduct(result, Def);
return result;
}
public static DistributionStructArray <TDist, T> ItemsAverageLogarithmInit <TDist>([IgnoreDependency] DistributionStructArray <TDist, T> array, IList <int> indices)
where TDist : struct,
SettableToProduct <TDist>,
SettableToRatio <TDist>,
SettableToPower <TDist>,
SettableToWeightedSum <TDist>,
CanGetLogAverageOf <TDist>,
CanGetLogAverageOfPower <TDist>,
CanGetAverageLog <TDist>,
IDistribution <T>,
Sampleable <T>
{
return(new DistributionStructArray <TDist, T>(indices.Count, i => (TDist)array[indices[i]].Clone()));
}
/// <summary>Computations that depend on the observed value of FeatureCount and FeatureValues and InstanceCount and Labels and numberOfIterations and WeightConstraints and WeightPriors</summary>
/// <param name="numberOfIterations">The number of times to iterate each loop</param>
private void Changed_FeatureCount_FeatureValues_InstanceCount_Labels_numberOfIterations_WeightConstraints_WeightP8(int numberOfIterations)
{
if (this.Changed_FeatureCount_FeatureValues_InstanceCount_Labels_numberOfIterations_WeightConstraints_WeightP8_isDone)
{
return;
}
for (int iteration = this.numberOfIterationsDone; iteration < numberOfIterations; iteration++)
{
for (int FeatureRange = 0; FeatureRange < this.FeatureCount; FeatureRange++)
{
this.Weights_depth1_rep_F_marginal[FeatureRange] = ReplicateOp_Divide.Marginal <Gaussian>(this.Weights_depth1_rep_B_toDef[FeatureRange], this.Weights_uses_F[1][FeatureRange], this.Weights_depth1_rep_F_marginal[FeatureRange]);
}
for (int InstanceRange = 0; InstanceRange < this.InstanceCount; InstanceRange++)
{
for (int FeatureRange = 0; FeatureRange < this.FeatureCount; FeatureRange++)
{
this.Weights_depth1_rep_F[FeatureRange][InstanceRange] = ReplicateOp_Divide.UsesAverageConditional <Gaussian>(this.Weights_depth1_rep_B[FeatureRange][InstanceRange], this.Weights_depth1_rep_F_marginal[FeatureRange], InstanceRange, this.Weights_depth1_rep_F[FeatureRange][InstanceRange]);
this.FeatureScores_F[InstanceRange][FeatureRange] = GaussianProductOpBase.ProductAverageConditional(this.FeatureValues[InstanceRange][FeatureRange], this.Weights_depth1_rep_F[FeatureRange][InstanceRange]);
}
this.Score_F[InstanceRange] = FastSumOp.SumAverageConditional(this.FeatureScores_F[InstanceRange]);
this.NoisyScore_F[InstanceRange] = GaussianFromMeanAndVarianceOp.SampleAverageConditional(this.Score_F[InstanceRange], 1.0);
this.NoisyScore_use_B[InstanceRange] = IsPositiveOp_Proper.XAverageConditional(Bernoulli.PointMass(this.Labels[InstanceRange]), this.NoisyScore_F[InstanceRange]);
this.Score_B[InstanceRange] = GaussianFromMeanAndVarianceOp.MeanAverageConditional(this.NoisyScore_use_B[InstanceRange], 1.0);
this.FeatureScores_B[InstanceRange] = FastSumOp.ArrayAverageConditional <DistributionStructArray <Gaussian, double> >(this.Score_B[InstanceRange], this.Score_F[InstanceRange], this.FeatureScores_F[InstanceRange], this.FeatureScores_B[InstanceRange]);
for (int FeatureRange = 0; FeatureRange < this.FeatureCount; FeatureRange++)
{
this.Weights_depth1_rep_B[FeatureRange][InstanceRange] = GaussianProductOpBase.BAverageConditional(this.FeatureScores_B[InstanceRange][FeatureRange], this.FeatureValues[InstanceRange][FeatureRange]);
this.Weights_depth1_rep_F_marginal[FeatureRange] = ReplicateOp_Divide.MarginalIncrement <Gaussian>(this.Weights_depth1_rep_F_marginal[FeatureRange], this.Weights_depth1_rep_F[FeatureRange][InstanceRange], this.Weights_depth1_rep_B[FeatureRange][InstanceRange]);
}
}
for (int FeatureRange = 0; FeatureRange < this.FeatureCount; FeatureRange++)
{
this.Weights_depth1_rep_B_toDef[FeatureRange] = ReplicateOp_Divide.ToDef <Gaussian>(this.Weights_depth1_rep_B[FeatureRange], this.Weights_depth1_rep_B_toDef[FeatureRange]);
}
this.OnProgressChanged(new ProgressChangedEventArgs(iteration));
}
for (int _iv = 0; _iv < this.FeatureCount; _iv++)
{
this.Weights_uses_B[1][_iv] = ArrayHelper.SetTo <Gaussian>(this.Weights_uses_B[1][_iv], this.Weights_depth1_rep_B_toDef[_iv]);
}
this.Weights_uses_F[0] = ReplicateOp_NoDivide.UsesAverageConditional <DistributionStructArray <Gaussian, double> >(this.Weights_uses_B, this.WeightPriors, 0, this.Weights_uses_F[0]);
this.ModelSelector_selector_cases_0_uses_B[6] = Bernoulli.FromLogOdds(ReplicateOp.LogEvidenceRatio <DistributionStructArray <Gaussian, double> >(this.Weights_uses_B, this.WeightPriors, this.Weights_uses_F));
this.ModelSelector_selector_cases_0_uses_B[7] = Bernoulli.FromLogOdds(ConstrainEqualRandomOp <double[]> .LogEvidenceRatio <DistributionStructArray <Gaussian, double> >(this.Weights_uses_F[0], this.WeightConstraints));
for (int FeatureRange = 0; FeatureRange < this.FeatureCount; FeatureRange++)
{
this.ModelSelector_selector_cases_0_rep3_uses_B[FeatureRange][1] = Bernoulli.FromLogOdds(ReplicateOp.LogEvidenceRatio <Gaussian>(this.Weights_depth1_rep_B[FeatureRange], this.Weights_uses_F[1][FeatureRange], this.Weights_depth1_rep_F[FeatureRange]));
this.ModelSelector_selector_cases_0_rep3_B[FeatureRange] = ReplicateOp_NoDivide.DefAverageConditional <Bernoulli>(this.ModelSelector_selector_cases_0_rep3_uses_B[FeatureRange], this.ModelSelector_selector_cases_0_rep3_B[FeatureRange]);
}
this.ModelSelector_selector_cases_0_uses_B[12] = ReplicateOp_NoDivide.DefAverageConditional <Bernoulli>(this.ModelSelector_selector_cases_0_rep3_B, this.ModelSelector_selector_cases_0_uses_B[12]);
for (int InstanceRange = 0; InstanceRange < this.InstanceCount; InstanceRange++)
{
this.ModelSelector_selector_cases_0_rep8_B[InstanceRange] = Bernoulli.FromLogOdds(IsPositiveOp.LogEvidenceRatio(this.Labels[InstanceRange], this.NoisyScore_F[InstanceRange]));
}
this.ModelSelector_selector_cases_0_uses_B[17] = ReplicateOp_NoDivide.DefAverageConditional <Bernoulli>(this.ModelSelector_selector_cases_0_rep8_B, this.ModelSelector_selector_cases_0_uses_B[17]);
this.ModelSelector_selector_cases_0_B = ReplicateOp_NoDivide.DefAverageConditional <Bernoulli>(this.ModelSelector_selector_cases_0_uses_B, this.ModelSelector_selector_cases_0_B);
this.ModelSelector_selector_cases_B[0] = ArrayHelper.SetTo <Bernoulli>(this.ModelSelector_selector_cases_B[0], this.ModelSelector_selector_cases_0_B);
this.ModelSelector_selector_B = CasesOp.BAverageConditional(this.ModelSelector_selector_cases_B);
this.ModelSelector_marginal_F = VariableOp.MarginalAverageConditional <Bernoulli>(this.ModelSelector_selector_B, this.vBernoulli0, this.ModelSelector_marginal_F);
this.Weights_use_B = ReplicateOp_NoDivide.DefAverageConditional <DistributionStructArray <Gaussian, double> >(this.Weights_uses_B, this.Weights_use_B);
this.Weights_marginal_F = VariableOp.MarginalAverageConditional <DistributionStructArray <Gaussian, double> >(this.Weights_use_B, this.WeightPriors, this.Weights_marginal_F);
this.Changed_FeatureCount_FeatureValues_InstanceCount_Labels_numberOfIterations_WeightConstraints_WeightP8_isDone = true;
}
/// <summary>Computations that depend on the observed value of Cloudy and WetGrass and Sprinkler</summary>
public void Changed_Cloudy_WetGrass_Sprinkler()
{
if (this.Changed_Cloudy_WetGrass_Sprinkler_iterationsDone == 1)
{
return;
}
// The constant 'vBernoulli0'
Bernoulli vBernoulli0 = Bernoulli.FromLogOdds(0);
this.Cloudy_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli0);
this.Cloudy_marginal = Distribution.SetPoint <Bernoulli, bool>(this.Cloudy_marginal, this.cloudy);
// The constant 'vBernoulli5'
Bernoulli vBernoulli5 = Bernoulli.FromLogOdds(4.5951198501345889);
this.WetGrass_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli5);
this.WetGrass_marginal = Distribution.SetPoint <Bernoulli, bool>(this.WetGrass_marginal, this.wetGrass);
// The constant 'vBernoulli1'
Bernoulli vBernoulli1 = Bernoulli.FromLogOdds(-2.1972245773362191);
this.Sprinkler_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
this.Sprinkler_marginal = Distribution.SetPoint <Bernoulli, bool>(this.Sprinkler_marginal, this.sprinkler);
// The constant 'vBernoulli3'
Bernoulli vBernoulli3 = Bernoulli.FromLogOdds(1.3862943611198908);
Bernoulli Rain_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
this.Rain_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message from use of 'Rain'
Bernoulli Rain_use_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
if (this.cloudy)
{
// Message to 'Rain' from Copy factor
Rain_F = Factor.Copy <Bernoulli>(vBernoulli3);
}
// The constant 'vBernoulli4'
Bernoulli vBernoulli4 = Bernoulli.FromLogOdds(-1.3862943611198906);
if (!this.cloudy)
{
// Message to 'Rain' from Copy factor
Rain_F = Factor.Copy <Bernoulli>(vBernoulli4);
}
// Rain_F is now updated in all contexts
if (this.sprinkler)
{
this.Rain_cond_Sprinkler_0_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_0' from Random factor
this.Rain_cond_Sprinkler_0_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli5));
// Create array for 'Rain_cond_Sprinkler_0_selector_cases' Backwards messages.
this.Rain_cond_Sprinkler_0_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
this.Rain_cond_Sprinkler_0_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_0_selector_cases_B[0] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_cases_0_B);
}
// The constant 'vBernoulli6'
Bernoulli vBernoulli6 = Bernoulli.FromLogOdds(2.1972245773362196);
if (this.sprinkler)
{
this.Rain_cond_Sprinkler_0_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_1' from Random factor
this.Rain_cond_Sprinkler_0_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_0_selector_cases_B[1] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_cases_1_B);
this.Rain_cond_Sprinkler_0_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_0_selector' from Cases factor
this.Rain_cond_Sprinkler_0_selector_B = CasesOp.BAverageConditional(this.Rain_cond_Sprinkler_0_selector_cases_B);
// Message to 'Rain_use' from Copy factor
Rain_use_B = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_B);
}
if (!this.sprinkler)
{
this.Rain_cond_Sprinkler_1_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_0' from Random factor
this.Rain_cond_Sprinkler_1_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
// Create array for 'Rain_cond_Sprinkler_1_selector_cases' Backwards messages.
this.Rain_cond_Sprinkler_1_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
this.Rain_cond_Sprinkler_1_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_1_selector_cases_B[0] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_cases_0_B);
}
// The constant 'vBernoulli8'
Bernoulli vBernoulli8 = Bernoulli.FromLogOdds(Double.NegativeInfinity);
if (!this.sprinkler)
{
this.Rain_cond_Sprinkler_1_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_1' from Random factor
this.Rain_cond_Sprinkler_1_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli8));
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_1_selector_cases_B[1] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_cases_1_B);
this.Rain_cond_Sprinkler_1_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_1_selector' from Cases factor
this.Rain_cond_Sprinkler_1_selector_B = CasesOp.BAverageConditional(this.Rain_cond_Sprinkler_1_selector_cases_B);
// Message to 'Rain_use' from Copy factor
Rain_use_B = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_B);
}
// Rain_use_B is now updated in all contexts
if (this.cloudy)
{
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Rain_use_B, Rain_F, this.Rain_marginal_F);
}
if (!this.cloudy)
{
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Rain_use_B, vBernoulli4, this.Rain_marginal_F);
}
// Rain_marginal_F is now updated in all contexts
this.Changed_Cloudy_WetGrass_Sprinkler_iterationsDone = 1;
}
/// <summary>Computations that depend on the observed value of Cloudy and WetGrass</summary>
public void Changed_Cloudy_WetGrass()
{
if (this.Changed_Cloudy_WetGrass_iterationsDone == 1)
{
return;
}
// The constant 'vBernoulli0'
Bernoulli vBernoulli0 = Bernoulli.FromLogOdds(0);
this.Cloudy_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli0);
this.Cloudy_marginal = Distribution.SetPoint <Bernoulli, bool>(this.Cloudy_marginal, this.cloudy);
// The constant 'vBernoulli5'
Bernoulli vBernoulli5 = Bernoulli.FromLogOdds(4.5951198501345889);
this.WetGrass_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli5);
this.WetGrass_marginal = Distribution.SetPoint <Bernoulli, bool>(this.WetGrass_marginal, this.wetGrass);
// The constant 'vBernoulli3'
Bernoulli vBernoulli3 = Bernoulli.FromLogOdds(1.3862943611198908);
Bernoulli Rain_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
this.Rain_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to use of 'Rain'
Bernoulli Rain_use_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message from use of 'Rain'
Bernoulli Rain_use_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
if (this.cloudy)
{
// Message to 'Rain' from Copy factor
Rain_F = Factor.Copy <Bernoulli>(vBernoulli3);
}
// The constant 'vBernoulli4'
Bernoulli vBernoulli4 = Bernoulli.FromLogOdds(-1.3862943611198906);
if (!this.cloudy)
{
// Message to 'Rain' from Copy factor
Rain_F = Factor.Copy <Bernoulli>(vBernoulli4);
}
// Rain_F is now updated in all contexts
if (this.cloudy)
{
// Message to 'Rain_use' from DerivedVariable factor
Rain_use_F = Factor.Copy <Bernoulli>(Rain_F);
}
Bernoulli Rain_cond_Sprinkler_0_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_0' from Random factor
Rain_cond_Sprinkler_0_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli5));
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_0_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler_0_selector_cases' Backwards messages.
Rain_cond_Sprinkler_0_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Rain_cond_Sprinkler_0_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
Rain_cond_Sprinkler_0_selector_cases_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_cases_0_B);
// The constant 'vBernoulli6'
Bernoulli vBernoulli6 = Bernoulli.FromLogOdds(2.1972245773362196);
Bernoulli Rain_cond_Sprinkler_0_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_1' from Random factor
Rain_cond_Sprinkler_0_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
Rain_cond_Sprinkler_0_selector_cases_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_cases_1_B);
if (!this.cloudy)
{
// Message to 'Rain_use' from DerivedVariable factor
Rain_use_F = Factor.Copy <Bernoulli>(vBernoulli4);
}
// Rain_use_F is now updated in all contexts
Bernoulli[] Sprinkler_selector_cases_0_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_cases_0_uses' Backwards messages.
Sprinkler_selector_cases_0_uses_B = new Bernoulli[5];
for (int _ind = 0; _ind < 5; _ind++)
{
Sprinkler_selector_cases_0_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Sprinkler_selector_cases_0_uses' from Cases factor
Sprinkler_selector_cases_0_uses_B[4] = Bernoulli.FromLogOdds(CasesOp.LogEvidenceRatio(Rain_cond_Sprinkler_0_selector_cases_B, Rain_use_F));
Bernoulli Sprinkler_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Sprinkler_selector_cases_0' from Replicate factor
Sprinkler_selector_cases_0_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_cases_0_uses_B, Sprinkler_selector_cases_0_B);
DistributionStructArray <Bernoulli, bool> Sprinkler_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Sprinkler_selector_cases' Backwards messages.
Sprinkler_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Sprinkler_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Sprinkler_selector_cases' from Copy factor
Sprinkler_selector_cases_B[0] = Factor.Copy <Bernoulli>(Sprinkler_selector_cases_0_B);
Bernoulli Rain_cond_Sprinkler_1_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_0' from Random factor
Rain_cond_Sprinkler_1_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_1_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler_1_selector_cases' Backwards messages.
Rain_cond_Sprinkler_1_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Rain_cond_Sprinkler_1_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
Rain_cond_Sprinkler_1_selector_cases_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_cases_0_B);
// The constant 'vBernoulli8'
Bernoulli vBernoulli8 = Bernoulli.FromLogOdds(Double.NegativeInfinity);
Bernoulli Rain_cond_Sprinkler_1_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_1' from Random factor
Rain_cond_Sprinkler_1_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli8));
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
Rain_cond_Sprinkler_1_selector_cases_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_cases_1_B);
Bernoulli[] Sprinkler_selector_cases_1_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_cases_1_uses' Backwards messages.
Sprinkler_selector_cases_1_uses_B = new Bernoulli[5];
for (int _ind = 0; _ind < 5; _ind++)
{
Sprinkler_selector_cases_1_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Sprinkler_selector_cases_1_uses' from Cases factor
Sprinkler_selector_cases_1_uses_B[4] = Bernoulli.FromLogOdds(CasesOp.LogEvidenceRatio(Rain_cond_Sprinkler_1_selector_cases_B, Rain_use_F));
Bernoulli Sprinkler_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Sprinkler_selector_cases_1' from Replicate factor
Sprinkler_selector_cases_1_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_cases_1_uses_B, Sprinkler_selector_cases_1_B);
// Message to 'Sprinkler_selector_cases' from Copy factor
Sprinkler_selector_cases_B[1] = Factor.Copy <Bernoulli>(Sprinkler_selector_cases_1_B);
Bernoulli[] Sprinkler_selector_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_uses' Backwards messages.
Sprinkler_selector_uses_B = new Bernoulli[2];
// The constant 'vBernoulli1'
Bernoulli vBernoulli1 = Bernoulli.FromLogOdds(-2.1972245773362191);
for (int _ind = 0; _ind < 2; _ind++)
{
Sprinkler_selector_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
}
// Message to 'Sprinkler_selector_uses' from Cases factor
Sprinkler_selector_uses_B[0] = CasesOp.BAverageConditional(Sprinkler_selector_cases_B);
Bernoulli Sprinkler_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
if (this.cloudy)
{
// Message to 'Sprinkler' from Copy factor
Sprinkler_F = Factor.Copy <Bernoulli>(vBernoulli1);
}
if (!this.cloudy)
{
// Message to 'Sprinkler' from Copy factor
Sprinkler_F = Factor.Copy <Bernoulli>(vBernoulli0);
}
// Sprinkler_F is now updated in all contexts
// Message to use of 'Sprinkler'
Bernoulli Sprinkler_use_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
if (this.cloudy)
{
// Message to 'Sprinkler_use' from DerivedVariable factor
Sprinkler_use_F = Factor.Copy <Bernoulli>(Sprinkler_F);
}
if (!this.cloudy)
{
// Message to 'Sprinkler_use' from DerivedVariable factor
Sprinkler_use_F = Factor.Copy <Bernoulli>(vBernoulli0);
}
// Sprinkler_use_F is now updated in all contexts
// Buffer for Replicate2BufferOp.UsesAverageConditional<Bernoulli>
Bernoulli Sprinkler_selector_uses_B_marginal = default(Bernoulli);
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_B_marginal = Replicate2BufferOp.MarginalInit <Bernoulli>(Sprinkler_use_F);
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_B_marginal = Replicate2BufferOp.Marginal <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_use_F, Sprinkler_selector_uses_B_marginal);
Bernoulli[] Sprinkler_selector_uses_F = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_uses' Forwards messages.
Sprinkler_selector_uses_F = new Bernoulli[2];
for (int _ind = 0; _ind < 2; _ind++)
{
Sprinkler_selector_uses_F[_ind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
}
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_F[1] = Replicate2BufferOp.UsesAverageConditional <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_use_F, Sprinkler_selector_uses_B_marginal, 1, Sprinkler_selector_uses_F[1]);
Bernoulli Rain_cond_Sprinkler_0_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_0_selector' from Cases factor
Rain_cond_Sprinkler_0_selector_B = CasesOp.BAverageConditional(Rain_cond_Sprinkler_0_selector_cases_B);
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler' Backwards messages.
Rain_cond_Sprinkler_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _gateind = 0; _gateind < 2; _gateind++)
{
Rain_cond_Sprinkler_B[_gateind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
}
// Message to 'Rain_cond_Sprinkler' from Copy factor
Rain_cond_Sprinkler_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_B);
Bernoulli Rain_cond_Sprinkler_1_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_1_selector' from Cases factor
Rain_cond_Sprinkler_1_selector_B = CasesOp.BAverageConditional(Rain_cond_Sprinkler_1_selector_cases_B);
// Message to 'Rain_cond_Sprinkler' from Copy factor
Rain_cond_Sprinkler_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_B);
// Message to 'Rain_use' from EnterPartial factor
Rain_use_B = GateEnterPartialOp <bool> .ValueAverageConditional <Bernoulli>(Rain_cond_Sprinkler_B, Sprinkler_selector_uses_F[1], Rain_use_F, new int[2] {
0, 1
}, Rain_use_B);
if (this.cloudy)
{
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Rain_use_B, Rain_F, this.Rain_marginal_F);
}
if (!this.cloudy)
{
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Rain_use_B, vBernoulli4, this.Rain_marginal_F);
}
// Rain_marginal_F is now updated in all contexts
this.Sprinkler_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
Bernoulli Sprinkler_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
// Message to 'Sprinkler_selector' from Replicate factor
Sprinkler_selector_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_selector_B);
if (this.cloudy)
{
// Message to 'Sprinkler_marginal' from DerivedVariable factor
this.Sprinkler_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Sprinkler_selector_B, Sprinkler_F, this.Sprinkler_marginal_F);
}
if (!this.cloudy)
{
// Message to 'Sprinkler_marginal' from DerivedVariable factor
this.Sprinkler_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Sprinkler_selector_B, vBernoulli0, this.Sprinkler_marginal_F);
}
// Sprinkler_marginal_F is now updated in all contexts
this.Changed_Cloudy_WetGrass_iterationsDone = 1;
}
/// <summary>
/// VMP Messages to Uses
/// </summary>
/// <param name="Uses">Nonconjugate messages from uses</param>
/// <param name="Def">Gaussian message from Defintion</param>
/// <param name="result">Previous message</param>
/// <returns></returns>
public static DistributionStructArray <Gaussian, double> UsesAverageLogarithm(DistributionStructArray <NonconjugateGaussian, double>[] Uses, DistributionStructArray <Gaussian, double> Def, DistributionStructArray <Gaussian, double> result)
{
// multiply the messages together
DistributionStructArray <NonconjugateGaussian, double> prod = Uses[0];
for (int i = 1; i < Uses.Length; i++)
{
prod.SetToProduct(prod, Uses[i]);
}
// convert to Gaussians
for (int i = 0; i < prod.Count; i++)
{
result[i] = prod[i].GetGaussian(true);
if ((!result[i].IsProper() && !result[i].IsUniform()) || double.IsNaN(result[i].Precision))
{
throw new ApplicationException("UsesAverageLogarithm ");
}
}
result.SetToProduct(result, Def);
return(result);
}
/// <summary>
/// VMP message to 'A'
/// </summary>
/// <param name="sum">Incoming message from 'Sum'. Must be a proper distribution. If uniform, the result will be uniform.</param>
/// <param name="A">Incoming message from 'A'. Must be a proper distribution. If all elements are uniform, the result will be uniform.</param>
/// <param name="B">Incoming message from 'B'. Must be a proper distribution. If any element is uniform, the result will be uniform.</param>
/// <param name="MeanOfB">Buffer 'MeanOfB'.</param>
/// <param name="CovarianceOfB">Buffer 'CovarianceOfB'.</param>
/// <param name="result">Modified to contain the outgoing message</param>
/// <returns><paramref name="result"/></returns>
/// <remarks><para>
/// The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except 'A'.
/// Because the factor is deterministic, 'Sum' is integrated out before taking the logarithm.
/// The formula is <c>exp(sum_(B) p(B) log(sum_Sum p(Sum) factor(Sum,A,B)))</c>.
/// </para></remarks>
/// <exception cref="ImproperMessageException"><paramref name="sum"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="A"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="B"/> is not a proper distribution</exception>
public static DistributionStructArray<Bernoulli, bool> AAverageLogarithm([SkipIfUniform] Gaussian sum, [Proper] DistributionStructArray<Bernoulli, bool> A, [Proper] VectorGaussian B, Vector MeanOfB, PositiveDefiniteMatrix CovarianceOfB, DistributionStructArray<Bernoulli, bool> result)
{
double mu = sum.GetMean();
var Ebbt = new PositiveDefiniteMatrix(A.Count, A.Count);
Ebbt.SetToSumWithOuter(CovarianceOfB, 1, MeanOfB, MeanOfB);
var ma = A.Select(x => x.GetMean()).ToArray();
for (int i = 0; i < A.Count; i++) {
double term1 = 0.0;
for (int j = 0; j < A.Count; j++)
if (j != i)
term1 += ma[i] * Ebbt[i, j];
var ri = result[i];
ri.LogOdds = -.5 * sum.Precision * (2.0 * term1 + Ebbt[i, i] - 2.0 * mu * MeanOfB[i]);
result[i] = ri;
}
return result;
}
/// <summary>
/// VMP message to 'Sum'
/// </summary>
/// <param name="A">Incoming message from 'A'.</param>
/// <param name="B">Constant value for 'B'.</param>
/// <returns>The outgoing VMP message to the 'Sum' argument</returns>
/// <remarks><para>
/// The outgoing message is a distribution matching the moments of 'Sum' as the random arguments are varied.
/// The formula is <c>proj[sum_(A) p(A) factor(Sum,A,B)]</c>.
/// </para></remarks>
public static Gaussian SumAverageLogarithm(DistributionStructArray<Bernoulli, bool> A, [SkipIfUniform] Vector B)
{
Gaussian result = new Gaussian();
// p(x|a,b) = N(E[a]'*E[b], E[b]'*var(a)*E[b] + E[a]'*var(b)*E[a] + trace(var(a)*var(b)))
Vector ma = Vector.Zero(A.Count);
Vector va = Vector.Zero(A.Count);
for (int i = 0; i < A.Count; i++) {
ma[i] = A[i].GetMean();
va[i] = A[i].GetVariance();
}
// Uses John Winn's rule for deterministic factors.
// Strict variational inference would set the variance to 0.
var MeanOfBSquared = Vector.Zero(B.Count);
MeanOfBSquared.SetToFunction(B, x => x * x);
result.SetMeanAndVariance(ma.Inner(B), va.Inner(MeanOfBSquared));
return result;
}
/// <summary>Computations that depend on the observed value of Sprinkler and WetGrass</summary>
public void Changed_Sprinkler_WetGrass()
{
if (this.Changed_Sprinkler_WetGrass_iterationsDone == 1)
{
return;
}
// The constant 'vBernoulli1'
Bernoulli vBernoulli1 = Bernoulli.FromLogOdds(-2.1972245773362191);
this.Sprinkler_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli1);
this.Sprinkler_marginal = Distribution.SetPoint <Bernoulli, bool>(this.Sprinkler_marginal, this.sprinkler);
// The constant 'vBernoulli5'
Bernoulli vBernoulli5 = Bernoulli.FromLogOdds(4.5951198501345889);
this.WetGrass_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli5);
this.WetGrass_marginal = Distribution.SetPoint <Bernoulli, bool>(this.WetGrass_marginal, this.wetGrass);
// The constant 'vBernoulli3'
Bernoulli vBernoulli3 = Bernoulli.FromLogOdds(1.3862943611198908);
Bernoulli Rain_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// The constant 'vBernoulli0'
Bernoulli vBernoulli0 = Bernoulli.FromLogOdds(0);
this.Cloudy_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli0);
Bernoulli[] Cloudy_selector_cases_0_uses_B = default(Bernoulli[]);
// Create array for 'Cloudy_selector_cases_0_uses' Backwards messages.
Cloudy_selector_cases_0_uses_B = new Bernoulli[7];
for (int _ind = 0; _ind < 7; _ind++)
{
Cloudy_selector_cases_0_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Cloudy_selector_cases_0_uses' from Random factor
Cloudy_selector_cases_0_uses_B[6] = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.sprinkler, vBernoulli1));
Bernoulli Cloudy_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Cloudy_selector_cases_0' from Replicate factor
Cloudy_selector_cases_0_B = ReplicateOp.DefAverageConditional <Bernoulli>(Cloudy_selector_cases_0_uses_B, Cloudy_selector_cases_0_B);
DistributionStructArray <Bernoulli, bool>[] Cloudy_selector_cases_uses_B = default(DistributionStructArray <Bernoulli, bool>[]);
// Create array for 'Cloudy_selector_cases_uses' Backwards messages.
Cloudy_selector_cases_uses_B = new DistributionStructArray <Bernoulli, bool> [3];
for (int _ind = 0; _ind < 3; _ind++)
{
// Create array for 'Cloudy_selector_cases_uses' Backwards messages.
Cloudy_selector_cases_uses_B[_ind] = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_uses_B[_ind][_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
}
// Message to 'Cloudy_selector_cases_uses' from Copy factor
Cloudy_selector_cases_uses_B[0][0] = Factor.Copy <Bernoulli>(Cloudy_selector_cases_0_B);
DistributionStructArray <Bernoulli, bool> Rain_cond_Cloudy_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Cloudy' Forwards messages.
Rain_cond_Cloudy_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _gateind = 0; _gateind < 2; _gateind++)
{
Rain_cond_Cloudy_F[_gateind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
}
// Message to 'Rain_cond_Cloudy' from Copy factor
Rain_cond_Cloudy_F[0] = Factor.Copy <Bernoulli>(vBernoulli3);
// The constant 'vBernoulli4'
Bernoulli vBernoulli4 = Bernoulli.FromLogOdds(-1.3862943611198906);
// Message to 'Rain_cond_Cloudy' from Copy factor
Rain_cond_Cloudy_F[1] = Factor.Copy <Bernoulli>(vBernoulli4);
if (this.sprinkler)
{
this.Rain_cond_Sprinkler_0_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_0' from Random factor
this.Rain_cond_Sprinkler_0_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli5));
// Create array for 'Rain_cond_Sprinkler_0_selector_cases' Backwards messages.
this.Rain_cond_Sprinkler_0_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
this.Rain_cond_Sprinkler_0_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_0_selector_cases_B[0] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_cases_0_B);
}
// The constant 'vBernoulli6'
Bernoulli vBernoulli6 = Bernoulli.FromLogOdds(2.1972245773362196);
if (this.sprinkler)
{
this.Rain_cond_Sprinkler_0_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_1' from Random factor
this.Rain_cond_Sprinkler_0_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_0_selector_cases_B[1] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_cases_1_B);
this.Rain_cond_Sprinkler_0_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_0_selector' from Cases factor
this.Rain_cond_Sprinkler_0_selector_B = CasesOp.BAverageConditional(this.Rain_cond_Sprinkler_0_selector_cases_B);
}
// Message from use of 'Rain'
Bernoulli Rain_use_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
if (this.sprinkler)
{
// Message to 'Rain_use' from Copy factor
Rain_use_B = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_0_selector_B);
}
if (!this.sprinkler)
{
this.Rain_cond_Sprinkler_1_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_0' from Random factor
this.Rain_cond_Sprinkler_1_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli6));
// Create array for 'Rain_cond_Sprinkler_1_selector_cases' Backwards messages.
this.Rain_cond_Sprinkler_1_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
this.Rain_cond_Sprinkler_1_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_1_selector_cases_B[0] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_cases_0_B);
}
// The constant 'vBernoulli8'
Bernoulli vBernoulli8 = Bernoulli.FromLogOdds(Double.NegativeInfinity);
if (!this.sprinkler)
{
this.Rain_cond_Sprinkler_1_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_1' from Random factor
this.Rain_cond_Sprinkler_1_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli8));
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
this.Rain_cond_Sprinkler_1_selector_cases_B[1] = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_cases_1_B);
this.Rain_cond_Sprinkler_1_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain_cond_Sprinkler_1_selector' from Cases factor
this.Rain_cond_Sprinkler_1_selector_B = CasesOp.BAverageConditional(this.Rain_cond_Sprinkler_1_selector_cases_B);
// Message to 'Rain_use' from Copy factor
Rain_use_B = Factor.Copy <Bernoulli>(this.Rain_cond_Sprinkler_1_selector_B);
}
// Rain_use_B is now updated in all contexts
// Message to 'Cloudy_selector_cases_uses' from Exit factor
Cloudy_selector_cases_uses_B[1] = GateExitOp <bool> .CasesAverageConditional <Bernoulli, DistributionStructArray <Bernoulli, bool> >(Rain_use_B, Rain_cond_Cloudy_F, Cloudy_selector_cases_uses_B[1]);
Bernoulli[] Cloudy_selector_cases_1_uses_B = default(Bernoulli[]);
// Create array for 'Cloudy_selector_cases_1_uses' Backwards messages.
Cloudy_selector_cases_1_uses_B = new Bernoulli[7];
for (int _ind = 0; _ind < 7; _ind++)
{
Cloudy_selector_cases_1_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Cloudy_selector_cases_1_uses' from Random factor
Cloudy_selector_cases_1_uses_B[6] = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.sprinkler, vBernoulli0));
Bernoulli Cloudy_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Cloudy_selector_cases_1' from Replicate factor
Cloudy_selector_cases_1_B = ReplicateOp.DefAverageConditional <Bernoulli>(Cloudy_selector_cases_1_uses_B, Cloudy_selector_cases_1_B);
// Message to 'Cloudy_selector_cases_uses' from Copy factor
Cloudy_selector_cases_uses_B[2][1] = Factor.Copy <Bernoulli>(Cloudy_selector_cases_1_B);
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Cloudy_selector_cases' Backwards messages.
Cloudy_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_B[_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Cloudy_selector_cases' from Replicate factor
Cloudy_selector_cases_B = ReplicateOp.DefAverageConditional <DistributionStructArray <Bernoulli, bool> >(Cloudy_selector_cases_uses_B, Cloudy_selector_cases_B);
Bernoulli Cloudy_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli0);
// Message to 'Cloudy_selector' from Cases factor
Cloudy_selector_B = CasesOp.BAverageConditional(Cloudy_selector_cases_B);
// Message to 'Cloudy_marginal' from Variable factor
this.Cloudy_marginal_F = VariableOp.MarginalAverageConditional <Bernoulli>(Cloudy_selector_B, vBernoulli0, this.Cloudy_marginal_F);
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Cloudy_selector_cases' Forwards messages.
Cloudy_selector_cases_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_F[_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
DistributionStructArray <Bernoulli, bool>[] Cloudy_selector_cases_uses_F = default(DistributionStructArray <Bernoulli, bool>[]);
// Create array for 'Cloudy_selector_cases_uses' Forwards messages.
Cloudy_selector_cases_uses_F = new DistributionStructArray <Bernoulli, bool> [3];
for (int _ind = 0; _ind < 3; _ind++)
{
// Create array for 'Cloudy_selector_cases_uses' Forwards messages.
Cloudy_selector_cases_uses_F[_ind] = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_uses_F[_ind][_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
}
// Message to 'Cloudy_selector_cases' from Cases factor
Cloudy_selector_cases_F = CasesOp.CasesAverageConditional <DistributionStructArray <Bernoulli, bool> >(vBernoulli0, Cloudy_selector_cases_F);
// Buffer for Replicate2BufferOp.UsesAverageConditional<DistributionStructArray<Bernoulli,bool>>
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_uses_B_marginal = default(DistributionStructArray <Bernoulli, bool>);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_B_marginal = Replicate2BufferOp.MarginalInit <DistributionStructArray <Bernoulli, bool> >(Cloudy_selector_cases_F);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_B_marginal = Replicate2BufferOp.Marginal <DistributionStructArray <Bernoulli, bool> >(Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_F[1] = Replicate2BufferOp.UsesAverageConditional <DistributionStructArray <Bernoulli, bool> >(Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal, 1, Cloudy_selector_cases_uses_F[1]);
this.Rain_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli3);
// Message to 'Rain' from Exit factor
Rain_F = GateExitOp <bool> .ExitAverageConditional <Bernoulli>(Rain_use_B, Cloudy_selector_cases_uses_F[1], Rain_cond_Cloudy_F, Rain_F);
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Rain_use_B, Rain_F, this.Rain_marginal_F);
this.Changed_Sprinkler_WetGrass_iterationsDone = 1;
}
/// <summary>
/// VMP message to 'A'
/// </summary>
/// <param name="sum">Incoming message from 'Sum'. Must be a proper distribution. If uniform, the result will be uniform.</param>
/// <param name="A">Incoming message from 'A'. Must be a proper distribution. If all elements are uniform, the result will be uniform.</param>
/// <param name="B">Constant value for 'B'.</param>
/// <param name="result">Modified to contain the outgoing message</param>
/// <returns><paramref name="result"/></returns>
/// <remarks><para>
/// The outgoing message is the factor viewed as a function of 'A' with 'Sum' integrated out.
/// The formula is <c>sum_Sum p(Sum) factor(Sum,A,B)</c>.
/// </para></remarks>
/// <exception cref="ImproperMessageException"><paramref name="sum"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="A"/> is not a proper distribution</exception>
public static DistributionStructArray <Bernoulli, bool> AAverageLogarithm([SkipIfUniform] Gaussian sum, [SkipIfAllUniform] DistributionStructArray <Bernoulli, bool> A, Vector B, DistributionStructArray <Bernoulli, bool> result)
{
double mu = sum.GetMean();
var ma = A.Select(x => x.GetMean()).ToArray();
for (int i = 0; i < A.Count; i++)
{
double term1 = 0.0;
for (int j = 0; j < A.Count; j++)
{
if (j != i)
{
term1 += ma[i] * B[i] * B[j];
}
}
var ri = result[i];
ri.LogOdds = -.5 * sum.Precision * (2.0 * term1 + B[i] * B[i] - 2.0 * mu * B[i]);
result[i] = ri;
}
return(result);
}
public static DistributionStructArray <Bernoulli, bool> AAverageLogarithm([SkipIfUniform] double sum, [SkipIfAllUniform] DistributionStructArray <Bernoulli, bool> A, Vector B)
{
throw new NotSupportedException(NotSupportedMessage);
}
/// <summary>
/// VMP message to 'A'
/// </summary>
/// <param name="sum">Incoming message from 'Sum'. Must be a proper distribution. If uniform, the result will be uniform.</param>
/// <param name="A">Incoming message from 'A'. Must be a proper distribution. If all elements are uniform, the result will be uniform.</param>
/// <param name="B">Incoming message from 'B'. Must be a proper distribution. If any element is uniform, the result will be uniform.</param>
/// <param name="MeanOfB">Buffer 'MeanOfB'.</param>
/// <param name="CovarianceOfB">Buffer 'CovarianceOfB'.</param>
/// <param name="result">Modified to contain the outgoing message</param>
/// <returns><paramref name="result"/></returns>
/// <remarks><para>
/// The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except 'A'.
/// Because the factor is deterministic, 'Sum' is integrated out before taking the logarithm.
/// The formula is <c>exp(sum_(B) p(B) log(sum_Sum p(Sum) factor(Sum,A,B)))</c>.
/// </para></remarks>
/// <exception cref="ImproperMessageException"><paramref name="sum"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="A"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="B"/> is not a proper distribution</exception>
public static DistributionStructArray <Bernoulli, bool> AAverageLogarithm([SkipIfUniform] Gaussian sum, [SkipIfAllUniform] DistributionStructArray <Bernoulli, bool> A, [SkipIfUniform] VectorGaussian B, Vector MeanOfB, PositiveDefiniteMatrix CovarianceOfB, DistributionStructArray <Bernoulli, bool> result)
{
double mu = sum.GetMean();
var Ebbt = new PositiveDefiniteMatrix(A.Count, A.Count);
Ebbt.SetToSumWithOuter(CovarianceOfB, 1, MeanOfB, MeanOfB);
var ma = A.Select(x => x.GetMean()).ToArray();
for (int i = 0; i < A.Count; i++)
{
double term1 = 0.0;
for (int j = 0; j < A.Count; j++)
{
if (j != i)
{
term1 += ma[i] * Ebbt[i, j];
}
}
var ri = result[i];
ri.LogOdds = -.5 * sum.Precision * (2.0 * term1 + Ebbt[i, i] - 2.0 * mu * MeanOfB[i]);
result[i] = ri;
}
return(result);
}
/// <summary>Computations that depend on the observed value of numberOfIterations and vint__0 and vint__1</summary>
/// <param name="numberOfIterations">The number of times to iterate each loop</param>
private void Changed_numberOfIterations_vint__0_vint__1(int numberOfIterations)
{
if (this.Changed_numberOfIterations_vint__0_vint__1_isDone)
{
return;
}
DistributionStructArray <Gaussian, double> vdouble__0_F;
Gaussian vdouble__0_F_reduced;
// Create array for 'vdouble__0' Forwards messages.
vdouble__0_F = new DistributionStructArray <Gaussian, double>(5);
// Message to 'vdouble__0' from GaussianFromMeanAndVariance factor
vdouble__0_F_reduced = GaussianFromMeanAndVarianceOp.SampleAverageConditional(6.0, 9.0);
for (int index1 = 0; index1 < 5; index1++)
{
vdouble__0_F[index1] = vdouble__0_F_reduced;
vdouble__0_F[index1] = vdouble__0_F_reduced;
}
// Create array for 'vdouble__0_marginal' Forwards messages.
this.vdouble__0_marginal_F = new DistributionStructArray <Gaussian, double>(5);
DistributionStructArray <Gaussian, double> vdouble__0_use_B;
// Create array for 'vdouble__0_use' Backwards messages.
vdouble__0_use_B = new DistributionStructArray <Gaussian, double>(5);
for (int index1 = 0; index1 < 5; index1++)
{
vdouble__0_use_B[index1] = Gaussian.Uniform();
}
DistributionStructArray <Gaussian, double>[] vdouble__0_uses_F;
// Create array for 'vdouble__0_uses' Forwards messages.
vdouble__0_uses_F = new DistributionStructArray <Gaussian, double> [2];
// Create array for 'vdouble__0_uses' Forwards messages.
vdouble__0_uses_F[1] = new DistributionStructArray <Gaussian, double>(5);
for (int index1 = 0; index1 < 5; index1++)
{
vdouble__0_uses_F[1][index1] = Gaussian.Uniform();
}
DistributionStructArray <Gaussian, double> vdouble__0_uses_F_1__marginal;
// Message to 'vdouble__0_itemvint__1_index0_' from GetItems factor
vdouble__0_uses_F_1__marginal = GetItemsOp <double> .MarginalInit <DistributionStructArray <Gaussian, double> >(vdouble__0_uses_F[1]);
DistributionStructArray <Gaussian, double> vdouble__0_itemvint__1_index0__F;
// Create array for 'vdouble__0_itemvint__1_index0_' Forwards messages.
vdouble__0_itemvint__1_index0__F = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble__0_itemvint__1_index0__F[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble11_F'
DistributionStructArray <Gaussian, double> vdouble11_F = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble11_F[index0] = Gaussian.Uniform();
}
// Create array for 'vdouble__0_uses' Forwards messages.
vdouble__0_uses_F[0] = new DistributionStructArray <Gaussian, double>(5);
for (int index1 = 0; index1 < 5; index1++)
{
vdouble__0_uses_F[0][index1] = Gaussian.Uniform();
}
DistributionStructArray <Gaussian, double> vdouble__0_uses_F_0__marginal;
// Message to 'vdouble__0_itemvint__0_index0_' from GetItems factor
vdouble__0_uses_F_0__marginal = GetItemsOp <double> .MarginalInit <DistributionStructArray <Gaussian, double> >(vdouble__0_uses_F[0]);
DistributionStructArray <Gaussian, double> vdouble__0_itemvint__0_index0__F;
// Create array for 'vdouble__0_itemvint__0_index0_' Forwards messages.
vdouble__0_itemvint__0_index0__F = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble__0_itemvint__0_index0__F[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble8_F'
DistributionStructArray <Gaussian, double> vdouble8_F = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble8_F[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble12_F'
DistributionStructArray <Gaussian, double> vdouble12_F = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble12_F[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble12_B'
DistributionStructArray <Gaussian, double> vdouble12_B = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble12_B[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble8_use_B'
DistributionStructArray <Gaussian, double> vdouble8_use_B = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble8_use_B[index0] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble11_use_B'
DistributionStructArray <Gaussian, double> vdouble11_use_B = new DistributionStructArray <Gaussian, double>(6);
for (int index0 = 0; index0 < 6; index0++)
{
vdouble11_use_B[index0] = Gaussian.Uniform();
}
for (int iteration = this.numberOfIterationsDone; iteration < numberOfIterations; iteration++)
{
// Message to 'vdouble__0_uses' from Replicate factor
vdouble__0_uses_F[1] = ReplicateOp_NoDivide.UsesAverageConditional <DistributionStructArray <Gaussian, double> >(this.vdouble__0_uses_B, vdouble__0_F, 1, vdouble__0_uses_F[1]);
// Message to 'vdouble__0_itemvint__1_index0_' from GetItems factor
vdouble__0_uses_F_1__marginal = GetItemsOp <double> .Marginal <DistributionStructArray <Gaussian, double>, Gaussian>(vdouble__0_uses_F[1], this.vdouble__0_uses_B[1], vdouble__0_uses_F_1__marginal);
// Message to 'vdouble__0_uses' from Replicate factor
vdouble__0_uses_F[0] = ReplicateOp_NoDivide.UsesAverageConditional <DistributionStructArray <Gaussian, double> >(this.vdouble__0_uses_B, vdouble__0_F, 0, vdouble__0_uses_F[0]);
// Message to 'vdouble__0_itemvint__0_index0_' from GetItems factor
vdouble__0_uses_F_0__marginal = GetItemsOp <double> .Marginal <DistributionStructArray <Gaussian, double>, Gaussian>(vdouble__0_uses_F[0], this.vdouble__0_uses_B[0], vdouble__0_uses_F_0__marginal);
for (int index0 = 0; index0 < 6; index0++)
{
// Message to 'vdouble__0_itemvint__1_index0_' from GetItems factor
vdouble__0_itemvint__1_index0__F[index0] = GetItemsOp <double> .ItemsAverageConditional <DistributionStructArray <Gaussian, double>, Gaussian>(this.vdouble__0_itemvint__1_index0__B[index0], vdouble__0_uses_F[1], vdouble__0_uses_F_1__marginal, this.Vint__1, index0, vdouble__0_itemvint__1_index0__F[index0]);
// Message to 'vdouble11' from GaussianFromMeanAndVariance factor
vdouble11_F[index0] = GaussianFromMeanAndVarianceOp.SampleAverageConditional(vdouble__0_itemvint__1_index0__F[index0], 1.0);
// Message to 'vdouble__0_itemvint__0_index0_' from GetItems factor
vdouble__0_itemvint__0_index0__F[index0] = GetItemsOp <double> .ItemsAverageConditional <DistributionStructArray <Gaussian, double>, Gaussian>(this.vdouble__0_itemvint__0_index0__B[index0], vdouble__0_uses_F[0], vdouble__0_uses_F_0__marginal, this.Vint__0, index0, vdouble__0_itemvint__0_index0__F[index0]);
// Message to 'vdouble8' from GaussianFromMeanAndVariance factor
vdouble8_F[index0] = GaussianFromMeanAndVarianceOp.SampleAverageConditional(vdouble__0_itemvint__0_index0__F[index0], 1.0);
// Message to 'vdouble12' from Difference factor
vdouble12_F[index0] = DoublePlusOp.AAverageConditional(vdouble8_F[index0], vdouble11_F[index0]);
// Message to 'vdouble12' from IsPositive factor
vdouble12_B[index0] = IsPositiveOp_Proper.XAverageConditional(Bernoulli.PointMass(true), vdouble12_F[index0]);
// Message to 'vdouble8_use' from Difference factor
vdouble8_use_B[index0] = DoublePlusOp.SumAverageConditional(vdouble12_B[index0], vdouble11_F[index0]);
// Message to 'vdouble__0_itemvint__0_index0_' from GaussianFromMeanAndVariance factor
this.vdouble__0_itemvint__0_index0__B[index0] = GaussianFromMeanAndVarianceOp.MeanAverageConditional(vdouble8_use_B[index0], 1.0);
// Message to 'vdouble11_use' from Difference factor
vdouble11_use_B[index0] = DoublePlusOp.BAverageConditional(vdouble8_F[index0], vdouble12_B[index0]);
// Message to 'vdouble__0_itemvint__1_index0_' from GaussianFromMeanAndVariance factor
this.vdouble__0_itemvint__1_index0__B[index0] = GaussianFromMeanAndVarianceOp.MeanAverageConditional(vdouble11_use_B[index0], 1.0);
}
// Message to 'vdouble__0_uses' from GetItems factor
this.vdouble__0_uses_B[0] = GetItemsOp <double> .ArrayAverageConditional <Gaussian, DistributionStructArray <Gaussian, double> >(this.vdouble__0_itemvint__0_index0__B, this.Vint__0, this.vdouble__0_uses_B[0]);
// Message to 'vdouble__0_uses' from GetItems factor
this.vdouble__0_uses_B[1] = GetItemsOp <double> .ArrayAverageConditional <Gaussian, DistributionStructArray <Gaussian, double> >(this.vdouble__0_itemvint__1_index0__B, this.Vint__1, this.vdouble__0_uses_B[1]);
this.OnProgressChanged(new ProgressChangedEventArgs(iteration));
}
// Message to 'vdouble__0_use' from Replicate factor
vdouble__0_use_B = ReplicateOp_NoDivide.DefAverageConditional <DistributionStructArray <Gaussian, double> >(this.vdouble__0_uses_B, vdouble__0_use_B);
for (int index1 = 0; index1 < 5; index1++)
{
this.vdouble__0_marginal_F[index1] = Gaussian.Uniform();
// Message to 'vdouble__0_marginal' from Variable factor
this.vdouble__0_marginal_F[index1] = VariableOp.MarginalAverageConditional <Gaussian>(vdouble__0_use_B[index1], vdouble__0_F_reduced, this.vdouble__0_marginal_F[index1]);
}
this.Changed_numberOfIterations_vint__0_vint__1_isDone = true;
}
/// <summary>
/// The marginal distribution is the same as the message to uses.
/// </summary>
/// <param name="Uses"></param>
/// <param name="Def"></param>
/// <param name="result"></param>
/// <returns></returns>
public static DistributionStructArray <Gaussian, double> MarginalAverageLogarithm(DistributionStructArray <NonconjugateGaussian, double>[] Uses, DistributionStructArray <Gaussian, double> Def, DistributionStructArray <Gaussian, double> result)
{
return(UsesAverageLogarithm(Uses, Def, result));
}
/// <summary>Computations that depend on the observed value of vVector__330 and vdouble__990</summary>
private void Changed_vVector__330_vdouble__990()
{
if (this.Changed_vVector__330_vdouble__990_iterationsDone == 1)
{
return;
}
this.vVector__330_marginal = new PointMass <Vector[]>(this.VVector__330);
this.vdouble__990_marginal = new DistributionStructArray <Gaussian, double>(5622, delegate(int index330) {
return(Gaussian.Uniform());
});
this.vdouble__990_marginal = Distribution.SetPoint <DistributionStructArray <Gaussian, double>, double[]>(this.vdouble__990_marginal, this.Vdouble__990);
// The constant 'vVectorGaussian330'
VectorGaussian vVectorGaussian330 = VectorGaussian.FromNatural(DenseVector.FromArray(new double[3] {
0.0, 0.0, 0.0
}), new PositiveDefiniteMatrix(new double[3, 3] {
{ 1.0, 0.0, 0.0 }, { 0.0, 1.0, 0.0 }, { 0.0, 0.0, 1.0 }
}));
this.vVector991_marginal_F = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian330);
// Message from use of 'vdouble__991'
DistributionStructArray <Gaussian, double> vdouble__991_use_B = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__991_use' Backwards messages.
vdouble__991_use_B = new DistributionStructArray <Gaussian, double>(5622);
for (int index330 = 0; index330 < 5622; index330++)
{
vdouble__991_use_B[index330] = Gaussian.Uniform();
// Message to 'vdouble__991_use' from GaussianFromMeanAndVariance factor
vdouble__991_use_B[index330] = GaussianFromMeanAndVarianceOp.MeanAverageConditional(this.Vdouble__990[index330], 0.1);
}
DistributionRefArray <VectorGaussian, Vector> vVector991_rep_B = default(DistributionRefArray <VectorGaussian, Vector>);
// Create array for 'vVector991_rep' Backwards messages.
vVector991_rep_B = new DistributionRefArray <VectorGaussian, Vector>(5622);
for (int index330 = 0; index330 < 5622; index330++)
{
vVector991_rep_B[index330] = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian330);
// Message to 'vVector991_rep' from InnerProduct factor
vVector991_rep_B[index330] = InnerProductOp.AAverageConditional(vdouble__991_use_B[index330], this.VVector__330[index330], vVector991_rep_B[index330]);
}
// Buffer for ReplicateOp_Divide.Marginal<VectorGaussian>
VectorGaussian vVector991_rep_B_toDef = default(VectorGaussian);
// Message to 'vVector991_rep' from Replicate factor
vVector991_rep_B_toDef = ReplicateOp_Divide.ToDefInit <VectorGaussian>(vVectorGaussian330);
// Message to 'vVector991_rep' from Replicate factor
vVector991_rep_B_toDef = ReplicateOp_Divide.ToDef <VectorGaussian>(vVector991_rep_B, vVector991_rep_B_toDef);
// Message to 'vVector991_marginal' from Variable factor
this.vVector991_marginal_F = VariableOp.MarginalAverageConditional <VectorGaussian>(vVector991_rep_B_toDef, vVectorGaussian330, this.vVector991_marginal_F);
DistributionStructArray <Gaussian, double> vdouble__991_F = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__991' Forwards messages.
vdouble__991_F = new DistributionStructArray <Gaussian, double>(5622);
for (int index330 = 0; index330 < 5622; index330++)
{
vdouble__991_F[index330] = Gaussian.Uniform();
}
DistributionRefArray <VectorGaussian, Vector> vVector991_rep_F = default(DistributionRefArray <VectorGaussian, Vector>);
// Create array for 'vVector991_rep' Forwards messages.
vVector991_rep_F = new DistributionRefArray <VectorGaussian, Vector>(5622);
for (int index330 = 0; index330 < 5622; index330++)
{
vVector991_rep_F[index330] = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian330);
}
// Buffer for ReplicateOp_Divide.UsesAverageConditional<VectorGaussian>
VectorGaussian vVector991_rep_F_marginal = default(VectorGaussian);
// Message to 'vVector991_rep' from Replicate factor
vVector991_rep_F_marginal = ReplicateOp_Divide.MarginalInit <VectorGaussian>(vVectorGaussian330);
// Message to 'vVector991_rep' from Replicate factor
vVector991_rep_F_marginal = ReplicateOp_Divide.Marginal <VectorGaussian>(vVector991_rep_B_toDef, vVectorGaussian330, vVector991_rep_F_marginal);
// Buffer for InnerProductOp.InnerProductAverageConditional
// Create array for replicates of 'vVector991_rep_F_index330__AMean'
Vector[] vVector991_rep_F_index330__AMean = new Vector[5622];
for (int index330 = 0; index330 < 5622; index330++)
{
// Message to 'vdouble__991' from InnerProduct factor
vVector991_rep_F_index330__AMean[index330] = InnerProductOp.AMeanInit(vVector991_rep_F[index330]);
}
// Buffer for InnerProductOp.AMean
// Create array for replicates of 'vVector991_rep_F_index330__AVariance'
PositiveDefiniteMatrix[] vVector991_rep_F_index330__AVariance = new PositiveDefiniteMatrix[5622];
for (int index330 = 0; index330 < 5622; index330++)
{
// Message to 'vdouble__991' from InnerProduct factor
vVector991_rep_F_index330__AVariance[index330] = InnerProductOp.AVarianceInit(vVector991_rep_F[index330]);
// Message to 'vVector991_rep' from Replicate factor
vVector991_rep_F[index330] = ReplicateOp_Divide.UsesAverageConditional <VectorGaussian>(vVector991_rep_B[index330], vVector991_rep_F_marginal, index330, vVector991_rep_F[index330]);
}
// Create array for 'vdouble__991_marginal' Forwards messages.
this.vdouble__991_marginal_F = new DistributionStructArray <Gaussian, double>(5622);
for (int index330 = 0; index330 < 5622; index330++)
{
this.vdouble__991_marginal_F[index330] = Gaussian.Uniform();
// Message to 'vdouble__991' from InnerProduct factor
vVector991_rep_F_index330__AVariance[index330] = InnerProductOp.AVariance(vVector991_rep_F[index330], vVector991_rep_F_index330__AVariance[index330]);
// Message to 'vdouble__991' from InnerProduct factor
vVector991_rep_F_index330__AMean[index330] = InnerProductOp.AMean(vVector991_rep_F[index330], vVector991_rep_F_index330__AVariance[index330], vVector991_rep_F_index330__AMean[index330]);
// Message to 'vdouble__991' from InnerProduct factor
vdouble__991_F[index330] = InnerProductOp.InnerProductAverageConditional(vVector991_rep_F_index330__AMean[index330], vVector991_rep_F_index330__AVariance[index330], this.VVector__330[index330]);
// Message to 'vdouble__991_marginal' from DerivedVariable factor
this.vdouble__991_marginal_F[index330] = DerivedVariableOp.MarginalAverageConditional <Gaussian>(vdouble__991_use_B[index330], vdouble__991_F[index330], this.vdouble__991_marginal_F[index330]);
}
this.Changed_vVector__330_vdouble__990_iterationsDone = 1;
}
/// <summary>
/// The marginal distribution is the same as the message to uses.
/// </summary>
/// <param name="Uses"></param>
/// <param name="Def"></param>
/// <param name="result"></param>
/// <returns></returns>
public static DistributionStructArray<Gaussian, double> MarginalAverageLogarithm(DistributionStructArray<NonconjugateGaussian, double>[] Uses, DistributionStructArray<Gaussian, double> Def, DistributionStructArray<Gaussian, double> result)
{
return UsesAverageLogarithm(Uses, Def, result);
}
/// <summary>Computations that depend on the observed value of vVector__251</summary>
private void Changed_vVector__251()
{
if (this.Changed_vVector__251_iterationsDone == 1)
{
return;
}
this.vVector__251_marginal = new PointMass <Vector[]>(this.VVector__251);
// The constant 'vVectorGaussian251'
VectorGaussian vVectorGaussian251 = VectorGaussian.FromNatural(DenseVector.FromArray(new double[3] {
1547829870.0, 525077980.0, 200270.0
}), new PositiveDefiniteMatrix(new double[3, 3] {
{ 4254590363351.0, 1127383488860.0, 433199230.0 }, { 1127383488860.0, 482723521821.0, 146764360.0 }, { 433199230.0, 146764360.0, 56221.0 }
}));
this.vVector753_marginal_F = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian251);
// Buffer for ReplicateOp_Divide.Marginal<VectorGaussian>
VectorGaussian vVector753_rep_B_toDef = default(VectorGaussian);
// Message to 'vVector753_rep' from Replicate factor
vVector753_rep_B_toDef = ReplicateOp_Divide.ToDefInit <VectorGaussian>(vVectorGaussian251);
// Message to 'vVector753_marginal' from Variable factor
this.vVector753_marginal_F = VariableOp.MarginalAverageConditional <VectorGaussian>(vVector753_rep_B_toDef, vVectorGaussian251, this.vVector753_marginal_F);
DistributionStructArray <Gaussian, double> vdouble__753_F = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__753' Forwards messages.
vdouble__753_F = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
vdouble__753_F[index251] = Gaussian.Uniform();
}
DistributionStructArray <Gaussian, double> vdouble__754_F = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__754' Forwards messages.
vdouble__754_F = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
vdouble__754_F[index251] = Gaussian.Uniform();
}
DistributionRefArray <VectorGaussian, Vector> vVector753_rep_F = default(DistributionRefArray <VectorGaussian, Vector>);
DistributionRefArray <VectorGaussian, Vector> vVector753_rep_B = default(DistributionRefArray <VectorGaussian, Vector>);
// Create array for 'vVector753_rep' Forwards messages.
vVector753_rep_F = new DistributionRefArray <VectorGaussian, Vector>(1);
// Create array for 'vVector753_rep' Backwards messages.
vVector753_rep_B = new DistributionRefArray <VectorGaussian, Vector>(1);
for (int index251 = 0; index251 < 1; index251++)
{
vVector753_rep_B[index251] = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian251);
vVector753_rep_F[index251] = ArrayHelper.MakeUniform <VectorGaussian>(vVectorGaussian251);
}
// Buffer for ReplicateOp_Divide.UsesAverageConditional<VectorGaussian>
VectorGaussian vVector753_rep_F_marginal = default(VectorGaussian);
// Message to 'vVector753_rep' from Replicate factor
vVector753_rep_F_marginal = ReplicateOp_Divide.MarginalInit <VectorGaussian>(vVectorGaussian251);
// Message to 'vVector753_rep' from Replicate factor
vVector753_rep_F_marginal = ReplicateOp_Divide.Marginal <VectorGaussian>(vVector753_rep_B_toDef, vVectorGaussian251, vVector753_rep_F_marginal);
// Buffer for InnerProductOp.InnerProductAverageConditional
// Create array for replicates of 'vVector753_rep_F_index251__AMean'
Vector[] vVector753_rep_F_index251__AMean = new Vector[1];
for (int index251 = 0; index251 < 1; index251++)
{
// Message to 'vdouble__754' from InnerProduct factor
vVector753_rep_F_index251__AMean[index251] = InnerProductOp.AMeanInit(vVector753_rep_F[index251]);
}
// Buffer for InnerProductOp.AMean
// Create array for replicates of 'vVector753_rep_F_index251__AVariance'
PositiveDefiniteMatrix[] vVector753_rep_F_index251__AVariance = new PositiveDefiniteMatrix[1];
for (int index251 = 0; index251 < 1; index251++)
{
// Message to 'vdouble__754' from InnerProduct factor
vVector753_rep_F_index251__AVariance[index251] = InnerProductOp.AVarianceInit(vVector753_rep_F[index251]);
// Message to 'vVector753_rep' from Replicate factor
vVector753_rep_F[index251] = ReplicateOp_Divide.UsesAverageConditional <VectorGaussian>(vVector753_rep_B[index251], vVector753_rep_F_marginal, index251, vVector753_rep_F[index251]);
}
// Create array for 'vdouble__754_marginal' Forwards messages.
this.vdouble__754_marginal_F = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
this.vdouble__754_marginal_F[index251] = Gaussian.Uniform();
}
// Message from use of 'vdouble__754'
DistributionStructArray <Gaussian, double> vdouble__754_use_B = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__754_use' Backwards messages.
vdouble__754_use_B = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
vdouble__754_use_B[index251] = Gaussian.Uniform();
// Message to 'vdouble__754' from InnerProduct factor
vVector753_rep_F_index251__AVariance[index251] = InnerProductOp.AVariance(vVector753_rep_F[index251], vVector753_rep_F_index251__AVariance[index251]);
// Message to 'vdouble__754' from InnerProduct factor
vVector753_rep_F_index251__AMean[index251] = InnerProductOp.AMean(vVector753_rep_F[index251], vVector753_rep_F_index251__AVariance[index251], vVector753_rep_F_index251__AMean[index251]);
// Message to 'vdouble__754' from InnerProduct factor
vdouble__754_F[index251] = InnerProductOp.InnerProductAverageConditional(vVector753_rep_F_index251__AMean[index251], vVector753_rep_F_index251__AVariance[index251], this.VVector__251[index251]);
// Message to 'vdouble__754_marginal' from DerivedVariable factor
this.vdouble__754_marginal_F[index251] = DerivedVariableOp.MarginalAverageConditional <Gaussian>(vdouble__754_use_B[index251], vdouble__754_F[index251], this.vdouble__754_marginal_F[index251]);
}
// Create array for 'vdouble__753_marginal' Forwards messages.
this.vdouble__753_marginal_F = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
this.vdouble__753_marginal_F[index251] = Gaussian.Uniform();
}
// Message from use of 'vdouble__753'
DistributionStructArray <Gaussian, double> vdouble__753_use_B = default(DistributionStructArray <Gaussian, double>);
// Create array for 'vdouble__753_use' Backwards messages.
vdouble__753_use_B = new DistributionStructArray <Gaussian, double>(1);
for (int index251 = 0; index251 < 1; index251++)
{
vdouble__753_use_B[index251] = Gaussian.Uniform();
// Message to 'vdouble__753' from GaussianFromMeanAndVariance factor
vdouble__753_F[index251] = GaussianFromMeanAndVarianceOp.SampleAverageConditional(vdouble__754_F[index251], 0.1);
// Message to 'vdouble__753_marginal' from Variable factor
this.vdouble__753_marginal_F[index251] = VariableOp.MarginalAverageConditional <Gaussian>(vdouble__753_use_B[index251], vdouble__753_F[index251], this.vdouble__753_marginal_F[index251]);
}
this.Changed_vVector__251_iterationsDone = 1;
}
/// <summary>Computations that depend on the observed value of numberOfIterationsDecreased and WetGrass</summary>
/// <param name="numberOfIterations">The number of times to iterate each loop</param>
public void Changed_numberOfIterationsDecreased_WetGrass(int numberOfIterations)
{
if (this.Changed_numberOfIterationsDecreased_WetGrass_iterationsDone == numberOfIterations)
{
return;
}
// The constant 'vBernoulli23'
Bernoulli vBernoulli23 = Bernoulli.FromLogOdds(4.5951198501345889);
this.WetGrass_marginal = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli23);
this.WetGrass_marginal = Distribution.SetPoint <Bernoulli, bool>(this.WetGrass_marginal, this.wetGrass);
Bernoulli Rain_F = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
// The constant 'vBernoulli18'
Bernoulli vBernoulli18 = Bernoulli.FromLogOdds(0);
this.Cloudy_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli18);
DistributionStructArray <Bernoulli, bool> Rain_cond_Cloudy_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Cloudy' Forwards messages.
Rain_cond_Cloudy_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _gateind = 0; _gateind < 2; _gateind++)
{
Rain_cond_Cloudy_F[_gateind] = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
}
// Message to 'Rain_cond_Cloudy' from Copy factor
Rain_cond_Cloudy_F[0] = Factor.Copy <Bernoulli>(this.vBernoulli21);
// The constant 'vBernoulli22'
Bernoulli vBernoulli22 = Bernoulli.FromLogOdds(0.28185115214098749);
// Message to 'Rain_cond_Cloudy' from Copy factor
Rain_cond_Cloudy_F[1] = Factor.Copy <Bernoulli>(vBernoulli22);
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Cloudy_selector_cases' Forwards messages.
Cloudy_selector_cases_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_F[_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Cloudy_selector_cases' from Cases factor
Cloudy_selector_cases_F = CasesOp.CasesAverageConditional <DistributionStructArray <Bernoulli, bool> >(vBernoulli18, Cloudy_selector_cases_F);
// The constant 'vBernoulli19'
Bernoulli vBernoulli19 = Bernoulli.FromLogOdds(-2.1972245773362191);
DistributionStructArray <Bernoulli, bool> Sprinkler_cond_Cloudy_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Sprinkler_cond_Cloudy' Forwards messages.
Sprinkler_cond_Cloudy_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _gateind = 0; _gateind < 2; _gateind++)
{
Sprinkler_cond_Cloudy_F[_gateind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
}
// Message to 'Sprinkler_cond_Cloudy' from Copy factor
Sprinkler_cond_Cloudy_F[0] = Factor.Copy <Bernoulli>(vBernoulli19);
// Message to 'Sprinkler_cond_Cloudy' from Copy factor
Sprinkler_cond_Cloudy_F[1] = Factor.Copy <Bernoulli>(vBernoulli18);
Bernoulli Rain_cond_Sprinkler_0_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_0' from Random factor
Rain_cond_Sprinkler_0_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli23));
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_0_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler_0_selector_cases' Backwards messages.
Rain_cond_Sprinkler_0_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Rain_cond_Sprinkler_0_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
Rain_cond_Sprinkler_0_selector_cases_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_cases_0_B);
// The constant 'vBernoulli24'
Bernoulli vBernoulli24 = Bernoulli.FromLogOdds(2.1972245773362196);
Bernoulli Rain_cond_Sprinkler_0_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_0_selector_cases_1' from Random factor
Rain_cond_Sprinkler_0_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli24));
// Message to 'Rain_cond_Sprinkler_0_selector_cases' from Copy factor
Rain_cond_Sprinkler_0_selector_cases_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_cases_1_B);
Bernoulli[] Sprinkler_selector_cases_0_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_cases_0_uses' Backwards messages.
Sprinkler_selector_cases_0_uses_B = new Bernoulli[5];
for (int _ind = 0; _ind < 5; _ind++)
{
Sprinkler_selector_cases_0_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
Bernoulli Sprinkler_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
DistributionStructArray <Bernoulli, bool> Sprinkler_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Sprinkler_selector_cases' Backwards messages.
Sprinkler_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Sprinkler_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
Bernoulli Rain_cond_Sprinkler_1_selector_cases_0_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_0' from Random factor
Rain_cond_Sprinkler_1_selector_cases_0_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli24));
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_1_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler_1_selector_cases' Backwards messages.
Rain_cond_Sprinkler_1_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _ind0 = 0; _ind0 < 2; _ind0++)
{
Rain_cond_Sprinkler_1_selector_cases_B[_ind0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
Rain_cond_Sprinkler_1_selector_cases_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_cases_0_B);
// The constant 'vBernoulli26'
Bernoulli vBernoulli26 = Bernoulli.FromLogOdds(-2.4423470353692043);
Bernoulli Rain_cond_Sprinkler_1_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
// Message to 'Rain_cond_Sprinkler_1_selector_cases_1' from Random factor
Rain_cond_Sprinkler_1_selector_cases_1_B = Bernoulli.FromLogOdds(UnaryOp <bool> .LogEvidenceRatio <Bernoulli>(this.wetGrass, vBernoulli26));
// Message to 'Rain_cond_Sprinkler_1_selector_cases' from Copy factor
Rain_cond_Sprinkler_1_selector_cases_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_cases_1_B);
Bernoulli[] Sprinkler_selector_cases_1_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_cases_1_uses' Backwards messages.
Sprinkler_selector_cases_1_uses_B = new Bernoulli[5];
for (int _ind = 0; _ind < 5; _ind++)
{
Sprinkler_selector_cases_1_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
Bernoulli Sprinkler_selector_cases_1_B = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
Bernoulli[] Sprinkler_selector_uses_B = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_uses' Backwards messages.
Sprinkler_selector_uses_B = new Bernoulli[2];
for (int _ind = 0; _ind < 2; _ind++)
{
Sprinkler_selector_uses_B[_ind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
}
Bernoulli Sprinkler_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
// Buffer for Replicate2BufferOp.UsesAverageConditional<DistributionStructArray<Bernoulli,bool>>
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_uses_B_marginal = default(DistributionStructArray <Bernoulli, bool>);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_B_marginal = Replicate2BufferOp.MarginalInit <DistributionStructArray <Bernoulli, bool> >(Cloudy_selector_cases_F);
DistributionStructArray <Bernoulli, bool>[] Cloudy_selector_cases_uses_F = default(DistributionStructArray <Bernoulli, bool>[]);
// Create array for 'Cloudy_selector_cases_uses' Forwards messages.
Cloudy_selector_cases_uses_F = new DistributionStructArray <Bernoulli, bool> [4];
for (int _ind = 0; _ind < 4; _ind++)
{
// Create array for 'Cloudy_selector_cases_uses' Forwards messages.
Cloudy_selector_cases_uses_F[_ind] = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_uses_F[_ind][_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
}
Bernoulli Sprinkler_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
// Buffer for Replicate2BufferOp.UsesAverageConditional<Bernoulli>
Bernoulli Sprinkler_selector_uses_B_marginal = default(Bernoulli);
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_B_marginal = Replicate2BufferOp.MarginalInit <Bernoulli>(Sprinkler_F);
Bernoulli[] Sprinkler_selector_uses_F = default(Bernoulli[]);
// Create array for 'Sprinkler_selector_uses' Forwards messages.
Sprinkler_selector_uses_F = new Bernoulli[2];
for (int _ind = 0; _ind < 2; _ind++)
{
Sprinkler_selector_uses_F[_ind] = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
}
Bernoulli Rain_cond_Sprinkler_0_selector_B = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
// Message to 'Rain_cond_Sprinkler_0_selector' from Cases factor
Rain_cond_Sprinkler_0_selector_B = CasesOp.BAverageConditional(Rain_cond_Sprinkler_0_selector_cases_B);
DistributionStructArray <Bernoulli, bool> Rain_cond_Sprinkler_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Rain_cond_Sprinkler' Backwards messages.
Rain_cond_Sprinkler_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _gateind = 0; _gateind < 2; _gateind++)
{
Rain_cond_Sprinkler_B[_gateind] = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
}
// Message to 'Rain_cond_Sprinkler' from Copy factor
Rain_cond_Sprinkler_B[0] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_0_selector_B);
Bernoulli Rain_cond_Sprinkler_1_selector_B = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
// Message to 'Rain_cond_Sprinkler_1_selector' from Cases factor
Rain_cond_Sprinkler_1_selector_B = CasesOp.BAverageConditional(Rain_cond_Sprinkler_1_selector_cases_B);
// Message to 'Rain_cond_Sprinkler' from Copy factor
Rain_cond_Sprinkler_B[1] = Factor.Copy <Bernoulli>(Rain_cond_Sprinkler_1_selector_B);
for (int iteration = this.Changed_numberOfIterationsDecreased_WetGrass_iterationsDone; iteration < numberOfIterations; iteration++)
{
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_B_marginal = Replicate2BufferOp.Marginal <DistributionStructArray <Bernoulli, bool> >(this.Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_F[1] = Replicate2BufferOp.UsesAverageConditional <DistributionStructArray <Bernoulli, bool> >(this.Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal, 1, Cloudy_selector_cases_uses_F[1]);
// Message to 'Rain' from Exit factor
Rain_F = GateExitOp <bool> .ExitAverageConditional <Bernoulli>(this.Rain_use_B, Cloudy_selector_cases_uses_F[1], Rain_cond_Cloudy_F, Rain_F);
// Message to 'Sprinkler_selector_cases_1_uses' from Cases factor
Sprinkler_selector_cases_1_uses_B[4] = Bernoulli.FromLogOdds(CasesOp.LogEvidenceRatio(Rain_cond_Sprinkler_1_selector_cases_B, Rain_F));
// Message to 'Sprinkler_selector_cases_1' from Replicate factor
Sprinkler_selector_cases_1_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_cases_1_uses_B, Sprinkler_selector_cases_1_B);
// Message to 'Sprinkler_selector_cases' from Copy factor
Sprinkler_selector_cases_B[1] = Factor.Copy <Bernoulli>(Sprinkler_selector_cases_1_B);
// Message to 'Sprinkler_selector_cases_0_uses' from Cases factor
Sprinkler_selector_cases_0_uses_B[4] = Bernoulli.FromLogOdds(CasesOp.LogEvidenceRatio(Rain_cond_Sprinkler_0_selector_cases_B, Rain_F));
// Message to 'Sprinkler_selector_cases_0' from Replicate factor
Sprinkler_selector_cases_0_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_cases_0_uses_B, Sprinkler_selector_cases_0_B);
// Message to 'Sprinkler_selector_cases' from Copy factor
Sprinkler_selector_cases_B[0] = Factor.Copy <Bernoulli>(Sprinkler_selector_cases_0_B);
// Message to 'Sprinkler_selector_uses' from Cases factor
Sprinkler_selector_uses_B[0] = CasesOp.BAverageConditional(Sprinkler_selector_cases_B);
// Message to 'Sprinkler_selector' from Replicate factor
Sprinkler_selector_B = ReplicateOp.DefAverageConditional <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_selector_B);
// Message to 'Cloudy_selector_cases_uses' from Exit factor
this.Cloudy_selector_cases_uses_B[3] = GateExitOp <bool> .CasesAverageConditional <Bernoulli, DistributionStructArray <Bernoulli, bool> >(Sprinkler_selector_B, Sprinkler_cond_Cloudy_F, this.Cloudy_selector_cases_uses_B[3]);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_B_marginal = Replicate2BufferOp.Marginal <DistributionStructArray <Bernoulli, bool> >(this.Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal);
// Message to 'Cloudy_selector_cases_uses' from Replicate factor
Cloudy_selector_cases_uses_F[3] = Replicate2BufferOp.UsesAverageConditional <DistributionStructArray <Bernoulli, bool> >(this.Cloudy_selector_cases_uses_B, Cloudy_selector_cases_F, Cloudy_selector_cases_uses_B_marginal, 3, Cloudy_selector_cases_uses_F[3]);
// Message to 'Sprinkler' from Exit factor
Sprinkler_F = GateExitOp <bool> .ExitAverageConditional <Bernoulli>(Sprinkler_selector_B, Cloudy_selector_cases_uses_F[3], Sprinkler_cond_Cloudy_F, Sprinkler_F);
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_B_marginal = Replicate2BufferOp.Marginal <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_F, Sprinkler_selector_uses_B_marginal);
// Message to 'Sprinkler_selector_uses' from Replicate factor
Sprinkler_selector_uses_F[1] = Replicate2BufferOp.UsesAverageConditional <Bernoulli>(Sprinkler_selector_uses_B, Sprinkler_F, Sprinkler_selector_uses_B_marginal, 1, Sprinkler_selector_uses_F[1]);
// Message to 'Rain_use' from EnterPartial factor
this.Rain_use_B = GateEnterPartialOp <bool> .ValueAverageConditional <Bernoulli>(Rain_cond_Sprinkler_B, Sprinkler_selector_uses_F[1], Rain_F, new int[2] {
0, 1
}, this.Rain_use_B);
// Message to 'Cloudy_selector_cases_uses' from Exit factor
this.Cloudy_selector_cases_uses_B[1] = GateExitOp <bool> .CasesAverageConditional <Bernoulli, DistributionStructArray <Bernoulli, bool> >(this.Rain_use_B, Rain_cond_Cloudy_F, this.Cloudy_selector_cases_uses_B[1]);
this.OnProgressChanged(new ProgressChangedEventArgs(iteration));
}
DistributionStructArray <Bernoulli, bool> Cloudy_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'Cloudy_selector_cases' Backwards messages.
Cloudy_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv0 = 0; _iv0 < 2; _iv0++)
{
Cloudy_selector_cases_B[_iv0] = ArrayHelper.MakeUniform <Bernoulli>(new Bernoulli());
}
// Message to 'Cloudy_selector_cases' from Replicate factor
Cloudy_selector_cases_B = ReplicateOp.DefAverageConditional <DistributionStructArray <Bernoulli, bool> >(this.Cloudy_selector_cases_uses_B, Cloudy_selector_cases_B);
Bernoulli Cloudy_selector_B = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli18);
// Message to 'Cloudy_selector' from Cases factor
Cloudy_selector_B = CasesOp.BAverageConditional(Cloudy_selector_cases_B);
// Message to 'Cloudy_marginal' from Variable factor
this.Cloudy_marginal_F = VariableOp.MarginalAverageConditional <Bernoulli>(Cloudy_selector_B, vBernoulli18, this.Cloudy_marginal_F);
this.Rain_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(this.vBernoulli21);
// Message to 'Rain_marginal' from DerivedVariable factor
this.Rain_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(this.Rain_use_B, Rain_F, this.Rain_marginal_F);
this.Sprinkler_marginal_F = ArrayHelper.MakeUniform <Bernoulli>(vBernoulli19);
// Message to 'Sprinkler_marginal' from DerivedVariable factor
this.Sprinkler_marginal_F = DerivedVariableOp.MarginalAverageConditional <Bernoulli>(Sprinkler_selector_B, Sprinkler_F, this.Sprinkler_marginal_F);
this.Changed_numberOfIterationsDecreased_WetGrass_iterationsDone = numberOfIterations;
}
/// <summary>
/// VMP message to 'B'
/// </summary>
/// <param name="sum">Incoming message from 'Sum'. Must be a proper distribution. If uniform, the result will be uniform.</param>
/// <param name="A">Incoming message from 'A'.</param>
/// <param name="result">Modified to contain the outgoing message</param>
/// <returns><paramref name="result"/></returns>
/// <remarks><para>
/// The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except 'B'.
/// Because the factor is deterministic, 'Sum' is integrated out before taking the logarithm.
/// The formula is <c>exp(sum_(A) p(A) log(sum_Sum p(Sum) factor(Sum,A,B)))</c>.
/// </para></remarks>
/// <exception cref="ImproperMessageException"><paramref name="sum"/> is not a proper distribution</exception>
public static VectorGaussian BAverageLogarithm([SkipIfUniform] Gaussian sum, DistributionStructArray<Bernoulli, bool> A, VectorGaussian result)
{
if (result == default(VectorGaussian)) result = new VectorGaussian(A.Count);
// E[log N(x; ab, 0)] = -0.5 E[(x-ab)^2]/0 = -0.5 (E[x^2] - 2 E[x] a' E[b] + trace(aa' E[bb']))/0
// message to a = N(a; E[x]*inv(var(b)+E[b]E[b]')*E[b], var(x)*inv(var(b)+E[b]E[b]'))
// result.Precision = (var(b)+E[b]*E[b]')/var(x)
// result.MeanTimesPrecision = E[x]/var(x)*E[b] = E[b]*X.MeanTimesPrecision
Vector ma = Vector.FromArray(A.Select(x => x.GetMean()).ToArray());
Vector va = Vector.FromArray(A.Select(x => x.GetVariance()).ToArray());
result.Precision.SetToDiagonal(va);
result.Precision.SetToSumWithOuter(result.Precision, 1, ma, ma);
result.Precision.SetToProduct(result.Precision, sum.Precision);
result.MeanTimesPrecision.SetToProduct(ma, sum.MeanTimesPrecision);
return result;
}
/// <summary>Computations that depend on the observed value of numberOfIterationsDecreased and vGamma__1 and vGaussian__1 and vDirichlet1</summary>
/// <param name="numberOfIterations">The number of times to iterate each loop</param>
private void Changed_numberOfIterationsDecreased_vGamma__1_vGaussian__1_vDirichlet1(int numberOfIterations)
{
if (this.Changed_numberOfIterationsDecreased_vGamma__1_vGaussian__1_vDirichlet1_iterationsDone == numberOfIterations)
{
return;
}
// Create array for 'vdouble__3_marginal' Forwards messages.
this.vdouble__3_marginal_F = new DistributionStructArray <Gaussian, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
this.vdouble__3_marginal_F[index2] = Gaussian.Uniform();
}
this.vVector1_marginal_F = ArrayHelper.MakeUniform <Dirichlet>(this.VDirichlet1);
Discrete vint9_F = ArrayHelper.MakeUniform <Discrete>(Discrete.Uniform(2));
Discrete[] vint9_selector_uses_B = default(Discrete[]);
// Create array for 'vint9_selector_uses' Backwards messages.
vint9_selector_uses_B = new Discrete[3];
for (int _ind = 0; _ind < 3; _ind++)
{
vint9_selector_uses_B[_ind] = ArrayHelper.MakeUniform <Discrete>(Discrete.Uniform(2));
}
DistributionStructArray <Bernoulli, bool>[] vint9_selector_cases_uses_B = default(DistributionStructArray <Bernoulli, bool>[]);
// Create array for 'vint9_selector_cases_uses' Backwards messages.
vint9_selector_cases_uses_B = new DistributionStructArray <Bernoulli, bool> [2];
for (int _ind = 0; _ind < 2; _ind++)
{
// Create array for 'vint9_selector_cases_uses' Backwards messages.
vint9_selector_cases_uses_B[_ind] = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv = 0; _iv < 2; _iv++)
{
vint9_selector_cases_uses_B[_ind][_iv] = Bernoulli.Uniform();
}
}
DistributionStructArray <Bernoulli, bool>[] vint9_selector_cases_depth1_uses_B = default(DistributionStructArray <Bernoulli, bool>[]);
// Create array for 'vint9_selector_cases_depth1_uses' Backwards messages.
vint9_selector_cases_depth1_uses_B = new DistributionStructArray <Bernoulli, bool> [2];
for (int _iv = 0; _iv < 2; _iv++)
{
// Create array for 'vint9_selector_cases_depth1_uses' Backwards messages.
vint9_selector_cases_depth1_uses_B[_iv] = new DistributionStructArray <Bernoulli, bool>(10);
for (int _ind = 0; _ind < 10; _ind++)
{
vint9_selector_cases_depth1_uses_B[_iv][_ind] = Bernoulli.Uniform();
}
}
// Create array for replicates of 'vdouble17__index2__F'
DistributionStructArray <Gaussian, double> vdouble17__index2__F = new DistributionStructArray <Gaussian, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
vdouble17__index2__F[index2] = Gaussian.Uniform();
}
// Message from use of 'vdouble17'
Gaussian vdouble17_use_B = Gaussian.Uniform();
// Message to marginal of 'vdouble17__index2_'
// Create array for replicates of 'vdouble17__index2__marginal_F'
DistributionStructArray <Gaussian, double> vdouble17__index2__marginal_F = new DistributionStructArray <Gaussian, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
vdouble17__index2__marginal_F[index2] = Gaussian.Uniform();
}
// Create array for replicates of 'vdouble__4_index2__B'
DistributionStructArray <Gamma, double> vdouble__4_index2__B = new DistributionStructArray <Gamma, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
vdouble__4_index2__B[index2] = Gamma.Uniform();
}
// Create array for 'vdouble__4_marginal' Forwards messages.
this.vdouble__4_marginal_F = new DistributionStructArray <Gamma, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
this.vdouble__4_marginal_F[index2] = Gamma.Uniform();
}
DistributionStructArray <Bernoulli, bool> vint9_selector_cases_depth1_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'vint9_selector_cases_depth1' Backwards messages.
vint9_selector_cases_depth1_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv = 0; _iv < 2; _iv++)
{
vint9_selector_cases_depth1_B[_iv] = Bernoulli.Uniform();
}
DistributionStructArray <Bernoulli, bool> vint9_selector_cases_B = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'vint9_selector_cases' Backwards messages.
vint9_selector_cases_B = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv = 0; _iv < 2; _iv++)
{
vint9_selector_cases_B[_iv] = Bernoulli.Uniform();
}
this.vint9_marginal_F = ArrayHelper.MakeUniform <Discrete>(Discrete.Uniform(2));
// Create array for replicates of 'vdouble__3_index2__B'
DistributionStructArray <Gaussian, double> vdouble__3_index2__B = new DistributionStructArray <Gaussian, double>(2);
for (int index2 = 0; index2 < 2; index2++)
{
vdouble__3_index2__B[index2] = Gaussian.Uniform();
// Message to 'vdouble__3_marginal' from Variable factor
this.vdouble__3_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gaussian>(this.vdouble__3_use_B[index2], this.VGaussian__1[index2], this.vdouble__3_marginal_F[index2]);
// Message to 'vdouble__4_marginal' from Variable factor
this.vdouble__4_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gamma>(this.vdouble__4_use_B[index2], this.VGamma__1[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble17__index2_' from Gaussian factor
vdouble17__index2__F[index2] = GaussianOp.SampleAverageLogarithm(this.vdouble__3_marginal_F[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble17__index2__marginal' from Variable factor
vdouble17__index2__marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gaussian>(vdouble17_use_B, vdouble17__index2__F[index2], vdouble17__index2__marginal_F[index2]);
}
// Message to 'vVector1_marginal' from Variable factor
this.vVector1_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Dirichlet>(this.vVector1_use_B, this.VDirichlet1, this.vVector1_marginal_F);
// Message to 'vint9' from Discrete factor
vint9_F = DiscreteFromDirichletOp.SampleAverageLogarithm(this.vVector1_marginal_F, vint9_F);
// Message to 'vint9_marginal' from Variable factor
this.vint9_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Discrete>(this.vint9_selector_B, vint9_F, this.vint9_marginal_F);
for (int iteration = this.Changed_numberOfIterationsDecreased_vGamma__1_vGaussian__1_vDirichlet1_iterationsDone; iteration < numberOfIterations; iteration++)
{
for (int index2 = 0; index2 < 2; index2++)
{
// Message to 'vdouble__4_marginal' from Variable factor
this.vdouble__4_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gamma>(this.vdouble__4_use_B[index2], this.VGamma__1[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble__3_index2_' from Gaussian factor
vdouble__3_index2__B[index2] = GaussianOp.MeanAverageLogarithm(vdouble17__index2__marginal_F[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble__3_use' from EnterOne factor
this.vdouble__3_use_B[index2] = GateEnterOneOp <double> .ValueAverageLogarithm <Gaussian>(vdouble__3_index2__B[index2], this.vint9_marginal_F, index2, this.vdouble__3_use_B[index2]);
// Message to 'vdouble__3_marginal' from Variable factor
this.vdouble__3_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gaussian>(this.vdouble__3_use_B[index2], this.VGaussian__1[index2], this.vdouble__3_marginal_F[index2]);
// Message to 'vdouble17__index2_' from Gaussian factor
vdouble17__index2__F[index2] = GaussianOp.SampleAverageLogarithm(this.vdouble__3_marginal_F[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble17__index2__marginal' from Variable factor
vdouble17__index2__marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gaussian>(vdouble17_use_B, vdouble17__index2__F[index2], vdouble17__index2__marginal_F[index2]);
// Message to 'vint9_selector_cases_depth1_uses' from Variable factor
vint9_selector_cases_depth1_uses_B[index2][8] = Bernoulli.FromLogOdds(VariableVmpOp.AverageLogFactor <Gaussian>(vdouble17__index2__marginal_F[index2]));
// Message to 'vint9_selector_cases_depth1_uses' from Gaussian factor
vint9_selector_cases_depth1_uses_B[index2][7] = Bernoulli.FromLogOdds(GaussianOp.AverageLogFactor(vdouble17__index2__marginal_F[index2], this.vdouble__3_marginal_F[index2], this.vdouble__4_marginal_F[index2]));
}
for (int _iv = 0; _iv < 2; _iv++)
{
// Message to 'vint9_selector_cases_depth1' from Replicate factor
vint9_selector_cases_depth1_B[_iv] = ReplicateOp.DefAverageLogarithm <Bernoulli>(vint9_selector_cases_depth1_uses_B[_iv], vint9_selector_cases_depth1_B[_iv]);
// Message to 'vint9_selector_cases_uses' from Copy factor
vint9_selector_cases_uses_B[0][_iv] = ArrayHelper.SetTo <Bernoulli>(vint9_selector_cases_uses_B[0][_iv], vint9_selector_cases_depth1_B[_iv]);
}
// Message to 'vint9_selector_cases' from Replicate factor
vint9_selector_cases_B = ReplicateOp.DefAverageLogarithm <DistributionStructArray <Bernoulli, bool> >(vint9_selector_cases_uses_B, vint9_selector_cases_B);
// Message to 'vint9_selector_uses' from CasesInt factor
vint9_selector_uses_B[0] = IntCasesOp.IAverageLogarithm(vint9_selector_cases_B, vint9_selector_uses_B[0]);
// Message to 'vint9_selector' from Replicate factor
this.vint9_selector_B = ReplicateOp.DefAverageLogarithm <Discrete>(vint9_selector_uses_B, this.vint9_selector_B);
// Message to 'vVector1_marginal' from Variable factor
this.vVector1_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Dirichlet>(this.vVector1_use_B, this.VDirichlet1, this.vVector1_marginal_F);
// Message to 'vint9' from Discrete factor
vint9_F = DiscreteFromDirichletOp.SampleAverageLogarithm(this.vVector1_marginal_F, vint9_F);
// Message to 'vint9_marginal' from Variable factor
this.vint9_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Discrete>(this.vint9_selector_B, vint9_F, this.vint9_marginal_F);
for (int index2 = 0; index2 < 2; index2++)
{
// Message to 'vdouble__4_index2_' from Gaussian factor
vdouble__4_index2__B[index2] = GaussianOp.PrecisionAverageLogarithm(vdouble17__index2__marginal_F[index2], this.vdouble__3_marginal_F[index2]);
// Message to 'vdouble__4_use' from EnterOne factor
this.vdouble__4_use_B[index2] = GateEnterOneOp <double> .ValueAverageLogarithm <Gamma>(vdouble__4_index2__B[index2], this.vint9_marginal_F, index2, this.vdouble__4_use_B[index2]);
}
// Message to 'vVector1_use' from Discrete factor
this.vVector1_use_B = DiscreteFromDirichletOp.ProbsAverageLogarithm(this.vint9_marginal_F, this.vVector1_use_B);
this.OnProgressChanged(new ProgressChangedEventArgs(iteration));
}
for (int index2 = 0; index2 < 2; index2++)
{
// Message to 'vdouble__3_index2_' from Gaussian factor
vdouble__3_index2__B[index2] = GaussianOp.MeanAverageLogarithm(vdouble17__index2__marginal_F[index2], this.vdouble__4_marginal_F[index2]);
// Message to 'vdouble__3_use' from EnterOne factor
this.vdouble__3_use_B[index2] = GateEnterOneOp <double> .ValueAverageLogarithm <Gaussian>(vdouble__3_index2__B[index2], this.vint9_marginal_F, index2, this.vdouble__3_use_B[index2]);
// Message to 'vdouble__3_marginal' from Variable factor
this.vdouble__3_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gaussian>(this.vdouble__3_use_B[index2], this.VGaussian__1[index2], this.vdouble__3_marginal_F[index2]);
// Message to 'vint9_selector_cases_depth1_uses' from Gaussian factor
vint9_selector_cases_depth1_uses_B[index2][7] = Bernoulli.FromLogOdds(GaussianOp.AverageLogFactor(vdouble17__index2__marginal_F[index2], this.vdouble__3_marginal_F[index2], this.vdouble__4_marginal_F[index2]));
}
for (int _iv = 0; _iv < 2; _iv++)
{
// Message to 'vint9_selector_cases_depth1' from Replicate factor
vint9_selector_cases_depth1_B[_iv] = ReplicateOp.DefAverageLogarithm <Bernoulli>(vint9_selector_cases_depth1_uses_B[_iv], vint9_selector_cases_depth1_B[_iv]);
// Message to 'vint9_selector_cases_uses' from Copy factor
vint9_selector_cases_uses_B[0][_iv] = ArrayHelper.SetTo <Bernoulli>(vint9_selector_cases_uses_B[0][_iv], vint9_selector_cases_depth1_B[_iv]);
}
// Message to 'vint9_selector_cases' from Replicate factor
vint9_selector_cases_B = ReplicateOp.DefAverageLogarithm <DistributionStructArray <Bernoulli, bool> >(vint9_selector_cases_uses_B, vint9_selector_cases_B);
// Message to 'vint9_selector_uses' from CasesInt factor
vint9_selector_uses_B[0] = IntCasesOp.IAverageLogarithm(vint9_selector_cases_B, vint9_selector_uses_B[0]);
// Message to 'vint9_selector' from Replicate factor
this.vint9_selector_B = ReplicateOp.DefAverageLogarithm <Discrete>(vint9_selector_uses_B, this.vint9_selector_B);
// Message to 'vint9_marginal' from Variable factor
this.vint9_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Discrete>(this.vint9_selector_B, vint9_F, this.vint9_marginal_F);
for (int index2 = 0; index2 < 2; index2++)
{
// Message to 'vdouble__4_index2_' from Gaussian factor
vdouble__4_index2__B[index2] = GaussianOp.PrecisionAverageLogarithm(vdouble17__index2__marginal_F[index2], this.vdouble__3_marginal_F[index2]);
// Message to 'vdouble__4_use' from EnterOne factor
this.vdouble__4_use_B[index2] = GateEnterOneOp <double> .ValueAverageLogarithm <Gamma>(vdouble__4_index2__B[index2], this.vint9_marginal_F, index2, this.vdouble__4_use_B[index2]);
// Message to 'vdouble__4_marginal' from Variable factor
this.vdouble__4_marginal_F[index2] = VariableVmpOp.MarginalAverageLogarithm <Gamma>(this.vdouble__4_use_B[index2], this.VGamma__1[index2], this.vdouble__4_marginal_F[index2]);
}
// Message to 'vVector1_use' from Discrete factor
this.vVector1_use_B = DiscreteFromDirichletOp.ProbsAverageLogarithm(this.vint9_marginal_F, this.vVector1_use_B);
// Message to 'vVector1_marginal' from Variable factor
this.vVector1_marginal_F = VariableVmpOp.MarginalAverageLogarithm <Dirichlet>(this.vVector1_use_B, this.VDirichlet1, this.vVector1_marginal_F);
this.vdouble17_F = Gaussian.Uniform();
DistributionStructArray <Bernoulli, bool> vint9_selector_cases_F = default(DistributionStructArray <Bernoulli, bool>);
// Create array for 'vint9_selector_cases' Forwards messages.
vint9_selector_cases_F = new DistributionStructArray <Bernoulli, bool>(2);
for (int _iv = 0; _iv < 2; _iv++)
{
vint9_selector_cases_F[_iv] = Bernoulli.Uniform();
}
// Message to 'vint9_selector_cases' from CasesInt factor
vint9_selector_cases_F = IntCasesOp.CasesAverageLogarithm <DistributionStructArray <Bernoulli, bool> >(this.vint9_marginal_F, vint9_selector_cases_F);
// Message to 'vdouble17' from Exit factor
this.vdouble17_F = GateExitOp <double> .ExitAverageLogarithm <Gaussian>(vint9_selector_cases_F, vdouble17__index2__marginal_F, this.vdouble17_F);
this.Changed_numberOfIterationsDecreased_vGamma__1_vGaussian__1_vDirichlet1_iterationsDone = numberOfIterations;
}
/// <summary>
/// VMP message to 'A'
/// </summary>
/// <param name="sum">Incoming message from 'Sum'. Must be a proper distribution. If uniform, the result will be uniform.</param>
/// <param name="A">Incoming message from 'A'. Must be a proper distribution. If all elements are uniform, the result will be uniform.</param>
/// <param name="B">Constant value for 'B'.</param>
/// <param name="result">Modified to contain the outgoing message</param>
/// <returns><paramref name="result"/></returns>
/// <remarks><para>
/// The outgoing message is the factor viewed as a function of 'A' with 'Sum' integrated out.
/// The formula is <c>sum_Sum p(Sum) factor(Sum,A,B)</c>.
/// </para></remarks>
/// <exception cref="ImproperMessageException"><paramref name="sum"/> is not a proper distribution</exception>
/// <exception cref="ImproperMessageException"><paramref name="A"/> is not a proper distribution</exception>
public static DistributionStructArray<Bernoulli, bool> AAverageLogarithm([SkipIfUniform] Gaussian sum, [Proper] DistributionStructArray<Bernoulli, bool> A, Vector B, DistributionStructArray<Bernoulli, bool> result)
{
double mu = sum.GetMean();
var ma = A.Select(x => x.GetMean()).ToArray();
for (int i = 0; i < A.Count; i++) {
double term1 = 0.0;
for (int j = 0; j < A.Count; j++)
if (j != i)
term1 += ma[i] * B[i] * B[j];
var ri = result[i];
ri.LogOdds = -.5 * sum.Precision * (2.0 * term1 + B[i]*B[i] - 2.0 * mu * B[i]);
result[i] = ri;
}
return result;
}
public static DistributionStructArray <Bernoulli, bool> AAverageLogarithm(double sum, [Proper] DistributionStructArray <Bernoulli, bool> A, Vector B)
{
throw new NotSupportedException(NotSupportedMessage);
}
