public static List <Evidence> Extend(List <Evidence> list, BayesianNode node, bool value)
{
return(new List <Evidence>(list)
{
new Evidence(node, value)
});
}
public void MakeDecision()
{
// You can specify a list of evidence
List <string> observations = new List <string> {
"brave=" + GetIsBrave(),
"enemy_amount=" + GetEnemyAmount(),
"cover_type=" + GetCoverType()
};
// You can then use them to infer another variable in the network
double[] fightDistribution = ve.Infer("fight", observations);
bool fight = ve.PickOne(fightDistribution) == 0;
// You can do chain interence based on previous inference results
observations.Add("fight=" + fight);
// The API functions are overloaded to fit your needs
// e.g. you can use a less string-based approach if you want to do things programmatically
BayesianNetwork network = ve.GetNetwork();
Proposition braveProp = network.FindNode("brave").Instantiate(GetIsBrave());
Proposition enemyAmountProp = network.FindNode("enemy_amount").Instantiate(GetEnemyAmount());
Proposition hasCoverProp = network.FindNode("cover_type").Instantiate(GetCoverType());
Proposition fightProp = network.FindNode("fight").Instantiate(fight.ToString());
BayesianNode runAwayNode = ve.GetNetwork().FindNode("run_away");
double[] runawayDistribution = ve.Infer(runAwayNode, braveProp, enemyAmountProp, hasCoverProp, fightProp);
bool runaway = ve.PickOne(runawayDistribution) == runAwayNode.var.GetTokenIndex("True");
// Since it is a bayesian network, you can infer any variables with partial or even no information
ve.Infer("enemy_amount", "fight=True");
ve.Infer("fight");
if (enemyAmount.Equals("NoEnemy"))
{
decisionText.text = "Did not see any enemy.";
}
else if (fight)
{
decisionText.text = "The NPC decided to fight. ";
}
else if (!fight && runaway)
{
decisionText.text = "The NPC decided to run away.";
}
else
{
decisionText.text = "The NPC decided to wait for his chance.";
}
decisionText.text = "Decision made: " + decisionText.text;
probabilityText.text = string.Format("true: {0}%\t\tfalse: {1}%\ntrue: {2}%\t\tfalse: {3}%",
fightDistribution[0] * 100, fightDistribution[1] * 100, runawayDistribution[0] * 100, runawayDistribution[1] * 100);
}
public Evidence(BayesianNode node, bool report)
{
Node = node;
Report = report;
}
public static List <BayesianNode> Rest(List <BayesianNode> list, BayesianNode first)
{
return(list.Where(n => n != first).ToList());
}
private void computarProbabilidadesToolStripMenuItem_Click(object sender, EventArgs e)
{
tabControl1.SelectTab(0);
try
{
if (CheckConsistency())
{
//obtener la evidencia
if (flowChartViewer1.Charts.Count > 0 && Evidence.Count > 0)
{
GetEvidence();
}
//transformar el grafo del visor a un grafo de nodos bayesianos
var graph = new BayesianGraph();
//
var dict = new Dictionary <BayesianNode, BayesianNodeChartElement>(flowChartViewer1.Charts.Count);
BayesianNode node;
var dict2 = new Dictionary <BayesianNodeChartElement, BayesianNode>(flowChartViewer1.Charts.Count);
graph.Clear();
foreach (var item in flowChartViewer1.Charts.Cast <BayesianNodeChartElement>())
{
node = new BayesianNode(item.Parents.Count, item.Childs.Count(), item.States.Count,
item.Parents.Select(x => x.States.Count).ToArray())
{
Probabilities = item.Condicional_Probabilities, Name = item.Name
};
dict.Add(node, item);
dict2.Add(item, node);
item.CreateConditionalProbabilities();
graph.AddVertex(node);
}
foreach (var bnodechart in flowChartViewer1.Charts.Cast <BayesianNodeChartElement>())
{
foreach (var child in bnodechart.Childs)
{
graph.AddEdge(dict2[bnodechart], dict2[child]);
}
}
//ejecutar el traspaso de mensajes en un poliarbol o en una red multiplemente conexa
progressBar1.Value = 5;
//if(MessageTransfer.ParalellStartNumber>graph.Count)
MessageTransfer.Increment += UpdateProgress;
if (graph.Arbol())
{
if (Evidence.Count > 0)
{
MessageTransfer.Message_Tranfer_Poliarbol(graph, Evidence[EvidenceIndex]);
}
else
{
MessageTransfer.Message_Tranfer_Poliarbol(graph);
}
foreach (var nodebayesian in graph)
{
dict[nodebayesian].SetConditionalProbabilities(nodebayesian.ConditionalProbability);
}
}
else
{
var markovnet = new MarkovNet(graph);
if (Evidence.Count > 0)
{
MessageTransfer.Message_Tranfer_Arbol_Union(markovnet, Evidence[EvidenceIndex]);
}
else
{
MessageTransfer.Message_Tranfer_Arbol_Union(markovnet);
}
foreach (var nodebayesian in markovnet.BayesianNet)
{
dict[nodebayesian].SetConditionalProbabilities(nodebayesian.ConditionalProbability);
}
}
timer1.Stop();
progressBar1.Value = 100;
Tests.Add(new BayesianTest()
{
Evidence = Evidence[EvidenceIndex].Clone(),
Probabilities =
graph.Select(x => x.ConditionalProbability.Clone() as double[]).ToArray()
});
TestsIndex = Tests.Count - 1;
ShowConditionalProbabilities();
}
}
catch (Exception ex)
{
MessageBox.Show("Hubo algun problema al calcular las probabilidades condicionales. " + ex.Message);
}
}
public Query(BayesianNode node, bool question)
{
Node = node;
Question = question;
}
/// <summary>
/// Se le llama al inicio de la escena
/// Se le llama cada vez que se pulsa un toogle o se desliza el slider
/// </summary>
public void MakeDecision()
{
////Lista con las observaciones de la escena
//List<string> observations = new List<string> {
// "brave=" + GetIsBrave(),
// "enemy_amount=" + GetEnemyAmount(),
// "cover_type=" + GetCoverType()
//};
///*EJEMPLO: [brave=True, enemy_amount= 3, cover_type=Full_cover] */
////Usamos la lista para inferir otra variable en la red
////Calculamos la probabilidad de luchar en función de la observación
//double[] fightDistribution = ve.Infer("fight", observations);
////Obtenemos si lucha o no
//bool fight = ve.PickOne(fightDistribution) == 0;
////Añadimos si lucha o no a la lista de observaciones para tenerlo en cuenta para las siguientes inferencias
//observations.Add("fight=" + fight);
// The API functions are overloaded to fit your needs
// e.g. you can use a less string-based approach if you want to do things programmatically
//Creamos la red de decisión
BayesianNetwork network = ve.GetNetwork();
//Obtenemos referencia a todas las proposiciones y las crea con los valores observables
Proposition braveProp = network.FindNode("brave").Instantiate(GetIsBrave());
Proposition enemyAmountProp = network.FindNode("enemy_amount").Instantiate(GetEnemyAmount());
Proposition hasCoverProp = network.FindNode("cover_type").Instantiate(GetCoverType());
BayesianNode fightNode = ve.GetNetwork().FindNode("fight");
double[] fightDistribution = ve.Infer(fightNode, braveProp, enemyAmountProp, hasCoverProp);
bool fight = ve.PickOne(fightDistribution) == fightNode.var.GetTokenIndex("True");
//Obtenemos la referencia a la proposición creada
Proposition fightProp = network.FindNode("fight").Instantiate(fight.ToString());
//Inferimos el nuevo nodo y tomamos una decisión
BayesianNode runAwayNode = ve.GetNetwork().FindNode("run_away");
double[] runawayDistribution = ve.Infer(runAwayNode, braveProp, enemyAmountProp, hasCoverProp, fightProp);
bool runaway = ve.PickOne(runawayDistribution) == runAwayNode.var.GetTokenIndex("True");
//Se puede inferir todo lo que quieras con información parcial o sin información
ve.Infer("enemy_amount", "fight=True");
ve.Infer("fight");
//Escribir la decisión en función de lo que nos salga del coñete
if (enemyAmount.Equals("NoEnemy"))
{
decisionText.text = "Did not see any enemy.";
}
else if (fight)
{
decisionText.text = "The NPC decided to fight. ";
}
else if (!fight && runaway)
{
decisionText.text = "The NPC decided to run away.";
}
else
{
decisionText.text = "The NPC decided to wait for his chance.";
}
decisionText.text = "Decision made: " + decisionText.text;
probabilityText.text = string.Format("true: {0}%\t\tfalse: {1}%\ntrue: {2}%\t\tfalse: {3}%",
fightDistribution[0] * 100, fightDistribution[1] * 100, runawayDistribution[0] * 100, runawayDistribution[1] * 100);
}