void TestSetDistributionSevenNodesModel()
{
BayesNet *net = SevenNodesModel();
if (net->GetGaussianMean("node0")[0].FltValue() != 0.5f)
{
PNL_THROW(pnl::CAlgorithmicException, "node0 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianMean("node1")[0].FltValue() != 0.5f)
{
PNL_THROW(pnl::CAlgorithmicException, "node1 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianCovar("node0")[0].FltValue() != 1.0f)
{
PNL_THROW(pnl::CAlgorithmicException, "node0 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianCovar("node1")[0].FltValue() != 1.0f)
{
PNL_THROW(pnl::CAlgorithmicException, "node1 : Setting or getting gaussian parameters is wrong");
}
float val12 = net->GetPTabular("node2")[0].FltValue();
float val22 = net->GetPTabular("node2")[1].FltValue();
if ((net->GetPTabular("node2")[0].FltValue() != 0.7f)||
(net->GetPTabular("node2")[1].FltValue() != 0.3f))
{
PNL_THROW(pnl::CAlgorithmicException, "node2 : Setting or getting tabular parameters is wrong");
};
TokArr off5True = net->GetSoftMaxOffset("node3", "node2^True");
if ((off5True[0].FltValue(0).fl != 0.3f)||
(off5True[0].FltValue(1).fl != 0.5f))
{
PNL_THROW(pnl::CAlgorithmicException, "node3 : Setting or getting softmax parameters is wrong");
};
TokArr off5False = net->GetSoftMaxOffset("node3", "node2^False");
float val1off = off5False[0].FltValue(0).fl;
float val2off = off5False[0].FltValue(1).fl;
if ((off5False[0].FltValue(0).fl != 0.3f)||
(off5False[0].FltValue(1).fl != 0.5f))
{
PNL_THROW(pnl::CAlgorithmicException, "node3 : Setting or getting softmax parameters is wrong");
};
TokArr node5True = net->GetSoftMaxWeights("node3", "node2^True");
if ((node5True[0].FltValue(0).fl != 0.5f)||
(node5True[0].FltValue(1).fl != 0.1f)||
(node5True[0].FltValue(2).fl != 0.5f)||
(node5True[0].FltValue(3).fl != 0.7f))
{
PNL_THROW(pnl::CAlgorithmicException, "node3 : Setting or getting softmax parameters is wrong");
};
TokArr node5False = net->GetSoftMaxWeights("node3", "node2^False");
float val0 = node5False[0].FltValue(0).fl;
float val1 = node5False[0].FltValue(1).fl;
float val2 = node5False[0].FltValue(2).fl;
float val3 = node5False[0].FltValue(3).fl;
if ((node5False[0].FltValue(0).fl != 0.5f)||
(node5False[0].FltValue(1).fl != 0.4f)||
(node5False[0].FltValue(2).fl != 0.5f)||
(node5False[0].FltValue(3).fl != 0.7f))
{
PNL_THROW(pnl::CAlgorithmicException, "node3 : Setting or getting softmax parameters is wrong");
};
float val40 = net->GetPTabular("node4", "node3^False")[0].FltValue();
float val41 = net->GetPTabular("node4", "node3^False")[1].FltValue();
float val42 = net->GetPTabular("node4", "node3^True")[0].FltValue() ;
float val43 = net->GetPTabular("node4", "node3^True")[1].FltValue() ;
if ((net->GetPTabular("node4", "node3^False")[0].FltValue() != 0.7f)||
(net->GetPTabular("node4", "node3^False")[1].FltValue() != 0.3f)||
(net->GetPTabular("node4", "node3^True")[0].FltValue() != 0.2f)||
(net->GetPTabular("node4", "node3^True")[1].FltValue() != 0.8f))
{
PNL_THROW(pnl::CAlgorithmicException, "node4 : Setting or getting tabular parameters is wrong");
};
if ((net->GetGaussianMean("node5", "node3^True")[0].FltValue() != 0.5f)||
(net->GetGaussianMean("node5", "node3^False")[0].FltValue() != 1.0f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
}
if ((net->GetGaussianCovar("node5", "node3^True")[0].FltValue() != 0.5f)||
(net->GetGaussianCovar("node5", "node3^False")[0].FltValue() != 1.0f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
}
TokArr off6True = net->GetSoftMaxOffset("node6", "node4^True");
if ((off6True[0].FltValue(0).fl != 0.1f)||
(off6True[0].FltValue(1).fl != 0.9f))
{
PNL_THROW(pnl::CAlgorithmicException, "node6 : Setting or getting softmax parameters is wrong");
};
TokArr off6False = net->GetSoftMaxOffset("node6", "node4^False");
if ((off6False[0].FltValue(0).fl != 0.7f)||
(off6False[0].FltValue(1).fl != 0.3f))
{
PNL_THROW(pnl::CAlgorithmicException, "node6 : Setting or getting softmax parameters is wrong");
};
TokArr node6True = net->GetSoftMaxWeights("node6", "node4^True");
if ((node6True[0].FltValue(0).fl != 0.8f)||
(node6True[0].FltValue(1).fl != 0.2f))
{
PNL_THROW(pnl::CAlgorithmicException, "node6 : Setting or getting softmax parameters is wrong");
};
TokArr node6False = net->GetSoftMaxWeights("node6", "node4^False");
float val06 = node5False[0].FltValue(0).fl;
float val16 = node5False[0].FltValue(1).fl;
if ((node6False[0].FltValue(0).fl != 0.5f)||
(node6False[0].FltValue(1).fl != 0.9f))
{
PNL_THROW(pnl::CAlgorithmicException, "node6 : Setting or getting softmax parameters is wrong");
};
delete net;
std::cout << "TestSetDistributionSevenNodesModel is completed successfully" << std::endl;
}
int main(int arg,char * argv[])
{
int a=1,b=2;
int c=a+b;
cout<<c<<endl;
//creating bayes net
//BayesNet net;
BayesNet net;
//adding node
net.AddNode("discrete^Cloudy","true false");
net.AddNode(discrete^"Sprinkler Rain WetGrass","true false");
//adding edges
net.AddArc("Cloudy","Sprinkler Rain");
net.AddArc("Sprinkler Rain","WetGrass");
//sopecfify the CPD
//cloudy
net.SetPTabular("Cloudy^true","0.6");
net.SetPTabular("Cloudy^false","0.4");
//spprinkler
net.SetPTabular("Sprinkler^true Sprinkler^false","0.1 0.9","Cloudy^true");
net.SetPTabular("Sprinkler^true Sprinkler^false","0.5 0.5","Cloudy^false");
//rain
net.SetPTabular("Rain^true Rain^false","0.8 0.2","Cloudy^true");
net.SetPTabular("Rain^true Rain^false","0.2 0.8","Cloudy^false");
//WetGrass
net.SetPTabular("WetGrass^true WetGrass^false","0.99 0.01","Sprinkler^true Rain^true");
net.SetPTabular("WetGrass^true WetGrass^false","0.9 0.1","Sprinkler^true Rain^false");
net.SetPTabular("WetGrass^true WetGrass^false","0.9 0.1","Sprinkler^false Rain^true");
net.SetPTabular("WetGrass^true WetGrass^false","0.0 1.0","Sprinkler^false Rain^false");
//get the cpd
TokArr PCloudy=net.GetPTabular("Cloudy");
String PCloudyStr=String(PCloudy);
float PCloudyTrueF=PCloudy[0].FltValue();
float PCloudyFalseF=PCloudy[1].FltValue();
cout<<endl<<"Cloudy"<<endl;
cout<<PCloudyStr<<endl;
cout<<PCloudyTrueF<<endl;
cout<<PCloudyFalseF<<endl;
/*
//adding evidence
//net.AddEvidToBuf("Rain^true WetGrass^true");
net.EditEvidence("Rain^true WetGrass^true");
net.CurEvidToBuf();
net.LearnParameters();
cout<<endl<<"evidence Rain^true WetGrass^true"<<endl;
//get the jpd
TokArr WetGrassMarg=net.GetJPD("WetGrass");
String WetGrassMargStr=String(WetGrassMarg);
cout<<endl<<"WetGrass JPD"<<endl<<WetGrassMargStr<<endl;
TokArr WetGrassAndSprinklerMarg=net.GetJPD("WetGrass Sprinkler Rain");
String WetGrassAndSprinklerMargStr=String(WetGrassAndSprinklerMarg);
cout<<endl<<"WetGrass and Sprinkler JPD"<<endl<<WetGrassAndSprinklerMargStr<<endl;
TokArr WetGrassMPE=net.GetMPE("WetGrass");
String WetGrassMPEStr=String(WetGrassMPE);
cout<<endl<<"WetGrass MPE"<<endl<<WetGrassMPEStr<<endl;
TokArr WetGrassAndSprinklerMPE=net.GetMPE("WetGrass Sprinkler Rain");
String WetGrassAndSprinklerMPEStr=String(WetGrassAndSprinklerMPE);
cout<<endl<<"WetGrass and Spinkler MPE"<<endl<<WetGrassAndSprinklerMPEStr<<endl;
//delete evidence
net.ClearEvid();
cout<<"ok"<<endl;*/
//net.AddEvidToBuf("Sprinkler^true WetGrass^true");
net.EditEvidence("Sprinkler^true WetGrass^true");
net.CurEvidToBuf();
net.LearnParameters();
cout<<endl<<"evidence Sprinkler^true WetGrass^true"<<endl;
//get jpd
TokArr WetGrassMarg=net.GetJPD("WetGrass");
String WetGrassMargStr=String(WetGrassMarg);
cout<<endl<<"WetGrass JPD"<<endl<<WetGrassMargStr<<endl;
TokArr WetGrassAndSprinklerMarg=net.GetJPD("WetGrass Sprinkler Rain");
String WetGrassAndSprinklerMargStr=String(WetGrassAndSprinklerMarg);
cout<<endl<<"WetGrass and Sprinkler JPD"<<endl<<WetGrassAndSprinklerMargStr<<endl;
TokArr WetGrassMPE=net.GetMPE("WetGrass");
String WetGrassMPEStr=String(WetGrassMPE);
cout<<endl<<"WetGrass MPE"<<endl<<WetGrassMPEStr<<endl;
TokArr WetGrassAndSprinklerMPE=net.GetMPE("WetGrass Sprinkler Rain Cloudy");
String WetGrassAndSprinklerMPEStr=String(WetGrassAndSprinklerMPE);
cout<<endl<<"WetGrass and Spinkler MPE"<<endl<<WetGrassAndSprinklerMPEStr<<endl;
cout<<endl<<"moonsea"<<endl;
return 0;
}
void TestSetDistributionCondSoftMax()
{
BayesNet *net = SimpleCondSoftMaxModel();
if (net->GetGaussianMean("node0")[0].FltValue() != 0.1f)
{
PNL_THROW(pnl::CAlgorithmicException, "node0 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianMean("node1")[0].FltValue() != 0.2f)
{
PNL_THROW(pnl::CAlgorithmicException, "node1 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianMean("node2")[0].FltValue() != 0.3f)
{
PNL_THROW(pnl::CAlgorithmicException, "node2 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianCovar("node0")[0].FltValue() != 0.9f)
{
PNL_THROW(pnl::CAlgorithmicException, "node0 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianCovar("node1")[0].FltValue() != 0.8f)
{
PNL_THROW(pnl::CAlgorithmicException, "node1 : Setting or getting gaussian parameters is wrong");
}
if (net->GetGaussianCovar("node2")[0].FltValue() != 0.7f)
{
PNL_THROW(pnl::CAlgorithmicException, "node2 : Setting or getting gaussian parameters is wrong");
}
if ((net->GetPTabular("node6")[0].FltValue() != 0.3f)||
(net->GetPTabular("node6")[1].FltValue() != 0.7f)||
(net->GetPTabular("node6")[2].FltValue() != 0.5f)||
(net->GetPTabular("node6")[3].FltValue() != 0.5f))
{
PNL_THROW(pnl::CAlgorithmicException, "node6 : Setting or getting gaussian parameters is wrong");
};
TokArr off5True = net->GetSoftMaxOffset("node5", "node3^True");
if ((off5True[0].FltValue(0).fl != 0.1f)||
(off5True[0].FltValue(1).fl != 0.1f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
};
TokArr off5False = net->GetSoftMaxOffset("node5", "node3^False");
float val1off = off5False[0].FltValue(0).fl;
float val2off = off5False[0].FltValue(1).fl;
if ((off5False[0].FltValue(0).fl != 0.21f)||
(off5False[0].FltValue(1).fl != 0.21f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
};
TokArr node5True = net->GetSoftMaxWeights("node5", "node3^True");
if ((node5True[0].FltValue(0).fl != 0.3f)||
(node5True[0].FltValue(1).fl != 0.4f)||
(node5True[0].FltValue(2).fl != 0.5f)||
(node5True[0].FltValue(3).fl != 0.6f)||
(node5True[0].FltValue(4).fl != 0.7f)||
(node5True[0].FltValue(5).fl != 0.8f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
};
TokArr node5False = net->GetSoftMaxWeights("node5", "node3^False");
float val0 = node5False[0].FltValue(0).fl;
float val1 = node5False[0].FltValue(1).fl;
float val2 = node5False[0].FltValue(2).fl;
float val3 = node5False[0].FltValue(3).fl;
float val4 = node5False[0].FltValue(4).fl;
float val5 = node5False[0].FltValue(5).fl;
if ((node5False[0].FltValue(0).fl != 0.23f)||
(node5False[0].FltValue(1).fl != 0.24f)||
(node5False[0].FltValue(2).fl != 0.25f)||
(node5False[0].FltValue(3).fl != 0.26f)||
(node5False[0].FltValue(4).fl != 0.27f)||
(node5False[0].FltValue(5).fl != 0.28f))
{
PNL_THROW(pnl::CAlgorithmicException, "node5 : Setting or getting gaussian parameters is wrong");
};
delete net;
std::cout << "TestSetDistributionCondSoftMax is completed successfully" << std::endl;
}
int main()
{
BayesNet net;
// adding nodes
net.AddNode("discrete^Cloudy", "true false");
net.AddNode("discrete^Sprinkler", "true false");
net.AddNode("discrete^Rain", "true false");
net.AddNode("discrete^WetGrass", "true false");
//adding edges
net.AddArc("Cloudy", "Sprinkler Rain");
net.AddArc("Sprinkler Rain", "WetGrass");
// specifying the conditional probabilities
net.SetPTabular("Cloudy^true Cloudy^false", "0.6 0.4");
net.SetPTabular("Sprinkler^true Sprinkler^false", "0.1 0.9", "Cloudy^true");
net.SetPTabular("Sprinkler^true Sprinkler^false", "0.5 0.5", "Cloudy^false");
net.SetPTabular("Rain^true Rain^false", "0.8 0.2", "Cloudy^true");
net.SetPTabular("Rain^true Rain^false", "0.2 0.8", "Cloudy^false");
//
net.SetPTabular("WetGrass^true WetGrass^false", "0.99 0.01", "Rain^true Sprinkler^true ");
net.SetPTabular("WetGrass^true WetGrass^false", "0.9 0.1", "Sprinkler^true Rain^false");
net.SetPTabular("WetGrass^true WetGrass^false", "0.9 0.1", "Sprinkler^false Rain^true");
net.SetPTabular("WetGrass^true WetGrass^false", "0.0 1.0", "Sprinkler^false Rain^false");
//To get the probability distribution of the node we must call the GetPTabular method:
TokArr PCloudy = net.GetPTabular("Cloudy");
// Now it is possible to represent this distribution as string or as float numbers:
String PCloudyStr = String(PCloudy);
float PCloudyTrueF = PCloudy[0].FltValue();
float PCloudyFalseF = PCloudy[1].FltValue();
cout << PCloudyStr << std::endl << PCloudyTrueF << "," << PCloudyFalseF << std::endl;
TokArr PSprinkler = net.GetPTabular("Sprinkler", "Cloudy^true");
String PSprinklerStr = String(PSprinkler);
float PSprinklerTrue = PSprinkler[0].FltValue();
float PSprinklerFalse = PSprinkler[1].FltValue();
cout << PSprinklerStr << std::endl << PSprinklerTrue << "," << PSprinklerFalse << std::endl;
// net.EditEvidence("Cloudy^false WetGrass^false");
// if the above line is un commented then after the net line the evidence buffer will have "Sprinkler^true Cloudy^true WetGrass^false"
net.EditEvidence("Sprinkler^true Cloudy^true");
TokArr PRain = net.GetJPD("Rain");
// Now it is possible to represent this distribution as string or as float numbers:
String PRainStr = String(PRain);
float PRainTrueF = PRain[0].FltValue();
float PRainFalseF = PRain[1].FltValue();
cout << PRainStr << std::endl << PRainTrueF << "," << PRainFalseF << std::endl;
TokArr PWetGrass = net.GetJPD("WetGrass");
String PWetGrassStr = String(PWetGrass);
float PWetGrassTrue = PWetGrass[0].FltValue();
float PWetGrassFalse = PWetGrass[1].FltValue();
cout << PWetGrassStr << std::endl << PWetGrassTrue << "," << PWetGrassFalse << std::endl;
return 0;
}