diff --git a/AMOEBAonAMAK/src/agents/context/Context.java b/AMOEBAonAMAK/src/agents/context/Context.java
index 2a4632fe95b9b884aaf0503014ff493fc76c79f8..fe93d333cfd63552b3211f0f4928cdc126763b9a 100644
--- a/AMOEBAonAMAK/src/agents/context/Context.java
+++ b/AMOEBAonAMAK/src/agents/context/Context.java
@@ -150,8 +150,7 @@ public class Context extends AmoebaAgent {
// world.trace(new ArrayList<String>(Arrays.asList(this.getName(),"NEW EXP",
// firstPoint.toString())));
- localModel.updateModel(this.getCurrentExperiment(), getAmas().data.learningSpeed,
- getAmas().data.numberOfPointsForRegression);
+ localModel.updateModel(this.getCurrentExperiment(), getAmas().data.learningSpeed);
getAmas().addAlteredContext(this);
this.setName(String.valueOf(this.hashCode()));
@@ -212,7 +211,7 @@ public class Context extends AmoebaAgent {
Double[] coef = ((LocalModelMillerRegression) fatherContext.localModel).getCoef();
((LocalModelMillerRegression) this.localModel).setCoef(coef);
this.actionProposition = ((LocalModelMillerRegression) fatherContext.localModel)
- .getProposition(fatherContext);
+ .getProposition();
}
getAmas().addAlteredContext(this);
@@ -274,13 +273,12 @@ public class Context extends AmoebaAgent {
Double[] coef = ((LocalModelMillerRegression) bestNearestContext.localModel).getCoef();
((LocalModelMillerRegression) this.localModel).setCoef(coef);
this.actionProposition = ((LocalModelMillerRegression) bestNearestContext.localModel)
- .getProposition(bestNearestContext);
+ .getProposition();
}
localModel.setFirstExperiments(new ArrayList<Experiment>(bestNearestContext.getLocalModel().getFirstExperiments()));
- localModel.updateModel(this.getCurrentExperiment(), getAmas().data.learningSpeed,
- getAmas().data.numberOfPointsForRegression);
+ localModel.updateModel(this.getCurrentExperiment(), getAmas().data.learningSpeed);
getAmas().addAlteredContext(this);
this.setName(String.valueOf(this.hashCode()));
@@ -446,37 +444,6 @@ public class Context extends AmoebaAgent {
return maxExpansions;
}
- // ---------
- // TODO these methods look very similar, maybe factorization is possible ?
- public void updateRequestNeighborState() { // faire le update dans le head attention partial et full
- if (nonValidNeightborPercepts.size() == 0) {
-
- getAmas().getHeadAgent().addRequestNeighbor(this);
- } else {
- getAmas().getHeadAgent().removeRequestNeighbor(this);
- }
- }
-
- public void updateActivatedContexts() { // faire le update dans le head attention partial et full
- if (nonValidPercepts.size() == 0) {
-
- getAmas().getHeadAgent().addActivatedContext(this);
- } else {
- getAmas().getHeadAgent().removeActivatedContext(this);
- }
- }
-
- public void updateActivatedContextsCopyForUpdate() { // faire le update dans le head attention partial et full
- if (nonValidPercepts.size() == 0) {
-
- getAmas().getHeadAgent().addActivatedContextCopy(this);
- } else {
- getAmas().getHeadAgent().removeActivatedContextCopy(this);
- }
-
- }
- // --------
-
public void clearNonValidPerceptNeighbors() {
nonValidNeightborPercepts.clear();
}
@@ -543,7 +510,7 @@ public class Context extends AmoebaAgent {
((LocalModelMillerRegression) this.localModel).setCoef(coef);
this.actionProposition = ((LocalModelMillerRegression) betterContext.getLocalModel())
- .getProposition(betterContext);
+ .getProposition();
}
public void analyzeResults3(Head head, Context closestContextToOracle) {
@@ -606,7 +573,7 @@ public class Context extends AmoebaAgent {
lastDistanceToModel = getLocalModel().distance(this.getCurrentExperiment());
lastAverageRegressionPerformanceIndicator = head.getAverageRegressionPerformanceIndicator();
- if(lastDistanceToModel < lastAverageRegressionPerformanceIndicator) {
+ if(lastDistanceToModel <= lastAverageRegressionPerformanceIndicator) {
//if(getLocalModel().distance(this.getCurrentExperiment()) < head.getAverageRegressionPerformanceIndicator()) {
//if (head.getCriticity(this) < head.getErrorAllowed()) {
confidence++;
@@ -959,14 +926,6 @@ public class Context extends AmoebaAgent {
return exp;
}
- private boolean tryNewExperiment2() {
- if (localModel.distance(getCurrentExperiment()) < 10.0) {
- localModel.updateModelWithExperimentAndWeight(getCurrentExperiment(), 0.5, 100);
- return true;
- }
- return false;
- }
-
public double sumOfRangesLengths() {
double sum = 0;
@@ -1821,56 +1780,6 @@ public class Context extends AmoebaAgent {
activations++;
getAmas().getHeadAgent().proposition(this);
}
- }
-
- private void onActOld() {
-
- if (nonValidPercepts.size() == 0) {
-
- getAmas().getHeadAgent().proposition(this);
-
-// for (Percept pct : getAmas().getPercepts()) {
-// getAmas().getHeadAgent().addPartiallyActivatedContextInNeighbors(pct, this);
-// }
-
- } else if (nonValidPercepts.size() == 1) {
- //getAmas().getHeadAgent().addPartiallyActivatedContext(nonValidPercepts.get(0), this);
- }
-
- if (nonValidNeightborPercepts.size() == 0) {
-
- getAmas().getHeadAgent().addRequestNeighbor(this);
- } else if (nonValidNeightborPercepts.size() == 1) {
- //getAmas().getHeadAgent().addPartialRequestNeighborContext(nonValidNeightborPercepts.get(0), this);
- }
-
- if ((nonValidNeightborPercepts.size() == 0) && (nonValidPercepts.size() == 1)) {
-
- //getAmas().getHeadAgent().addPartiallyActivatedContextInNeighbors(nonValidPercepts.get(0), this);
-
- }
-
- this.activations = 0;
- //this.valid = false;
-
- // Reset percepts validities
- for (Percept percept : perceptValidities.keySet()) {
- perceptValidities.put(percept, false);
- perceptNeighborhoodValidities.put(percept, false);
- }
-
- }
-
- public void computeContextNeighborsValidity() {
-
-
- if (nonValidNeightborPercepts.size() == 1) {
- getAmas().getHeadAgent().addPartiallyActivatedContextInNeighbors(nonValidNeightborPercepts.get(0), this);
- }
-
-
-
-
}
/**
@@ -1910,8 +1819,8 @@ public class Context extends AmoebaAgent {
s += "Context : " + getName() + "\n";
s += "Model : ";
s += this.localModel.getCoefsFormula() + "\n";
- s += "Max Prediction " + getLocalModel().getMaxProposition(this) + "\n";
- s += "Min Prediction " + getLocalModel().getMinProposition(this) + "\n";
+ s += "Max Prediction " + getLocalModel().getMaxProposition() + "\n";
+ s += "Min Prediction " + getLocalModel().getMinProposition() + "\n";
return s;
}
@@ -1922,8 +1831,8 @@ public class Context extends AmoebaAgent {
for(int i =1;i<localModel.getCoef().length;i++) {
array.add(""+localModel.getCoef()[i]);
}
- array.add(""+ getLocalModel().getMinProposition(this));
- array.add(""+ getLocalModel().getMaxProposition(this));
+ array.add(""+ getLocalModel().getMinProposition());
+ array.add(""+ getLocalModel().getMaxProposition());
return array;
}
@@ -1951,8 +1860,8 @@ public class Context extends AmoebaAgent {
s += "Mean Distance To Regression " + criticalities.getCriticalityMean("distanceToRegression") + "\n";
s += "Distance To Regression Allowed " + regressionPerformance.getPerformanceIndicator() +"\n\n";
- s += "Max Prediction " + getLocalModel().getMaxProposition(this) + "\n";
- s += "Min Prediction " + getLocalModel().getMinProposition(this) + "\n\n";
+ s += "Max Prediction " + getLocalModel().getMaxProposition() + "\n";
+ s += "Min Prediction " + getLocalModel().getMinProposition() + "\n\n";
s += "ASKED REQUEST " + waitingRequests.size() + "\n";
for(EndogenousRequest rqt : waitingRequests) {
@@ -2076,7 +1985,7 @@ public class Context extends AmoebaAgent {
}
public double getActionProposal() {
- return localModel.getProposition(this);
+ return localModel.getProposition();
}
public double getCenterByPercept(Percept pct) {
@@ -2133,7 +2042,7 @@ public class Context extends AmoebaAgent {
*/
public void setLocalModel(LocalModel localModel) {
this.localModel = localModel;
- this.localModel.context = this;
+ this.localModel.setContext(this);
}
diff --git a/AMOEBAonAMAK/src/agents/context/Range.java b/AMOEBAonAMAK/src/agents/context/Range.java
index 082754b311e52a8b0eab6b7606df584eb60278c6..61b53491ba423c369b008b839cc336400ddf5f7b 100644
--- a/AMOEBAonAMAK/src/agents/context/Range.java
+++ b/AMOEBAonAMAK/src/agents/context/Range.java
@@ -1484,8 +1484,6 @@ public class Range implements Serializable, Comparable, Cloneable {
}
}
- this.context.updateRequestNeighborState();
-
if (!this.contains2(percept.getValue())) {
if (!this.context.getNonValidPercepts().contains(this.percept)) {
this.context.addNonValidPercept(this.percept);
@@ -1496,10 +1494,6 @@ public class Range implements Serializable, Comparable, Cloneable {
}
}
- this.context.updateActivatedContextsCopyForUpdate();
-
-
-
if (!this.context.isDying() && !context.getAmas().getSpatiallyAlteredContextForUnityUI().contains(this.context)) {
context.getAmas().addSpatiallyAlteredContextForUnityUI(this.context);
}
@@ -1548,8 +1542,6 @@ public class Range implements Serializable, Comparable, Cloneable {
}
}
- this.context.updateRequestNeighborState();
-
if (!this.contains2(percept.getValue())) {
if (!this.context.getNonValidPercepts().contains(this.percept)) {
this.context.addNonValidPercept(this.percept);
@@ -1560,10 +1552,6 @@ public class Range implements Serializable, Comparable, Cloneable {
}
}
- this.context.updateActivatedContextsCopyForUpdate();
-
-
-
if (!this.context.isDying() && !context.getAmas().getSpatiallyAlteredContextForUnityUI().contains(this.context)) {
context.getAmas().addSpatiallyAlteredContextForUnityUI(this.context);
}
diff --git a/AMOEBAonAMAK/src/agents/context/localModel/LocalModel.java b/AMOEBAonAMAK/src/agents/context/localModel/LocalModel.java
index a2268693f46e26bb676a1f63ed7c56b8a7ba20db..2aab80c87bae8e396c3e7eaec30e4073612fb3b9 100644
--- a/AMOEBAonAMAK/src/agents/context/localModel/LocalModel.java
+++ b/AMOEBAonAMAK/src/agents/context/localModel/LocalModel.java
@@ -1,55 +1,110 @@
package agents.context.localModel;
import java.util.ArrayList;
+import java.util.HashMap;
import agents.context.Context;
import agents.context.Experiment;
-import agents.percept.Percept;
/**
- * The abstract class of all agents in charge of the generation of the output
- * from Context Agent. For the sake of simplicity, it's not scheduled as agent
- * like other of the system.
+ * A LocalModel is used by a Context to store information and generate prediction.
*/
-public abstract class LocalModel {
-
- public Context context;
+public interface LocalModel {
/**
- * Instantiates a new local model agent.
+ * Sets the context that use the LocalModel
+ * @param context
*/
- public LocalModel(Context associatedContext) {
- context = associatedContext;
- }
-
+ public void setContext(Context context);
+
+ /**
+ * gets the context that use the LocalModel
+ * @return
+ */
+ public Context getContext();
+
/**
* Gets the proposition.
*
- * @param context the context
* @return the proposition
*/
- public abstract double getProposition(Context context);
- public abstract double getProposition(Experiment experiment);
- public abstract double getMaxProposition(Context context);
- public abstract double getMinProposition(Context context);
+ public double getProposition();
-
+ /**
+ * Gets the proposition with the highest value possible
+ * @return
+ */
+ public double getMaxProposition();
+
+ /**
+ * Return the point (percept value) that produce the max proposition, considering some percepts are fixed.
+ * @return a HashMap with percept names as key, and their corresponding value. The oracle is the max proposition
+ * @see LocalModel#getMaxProposition(Context)
+ */
+ public HashMap<String, Double> getMaxWithConstraint(HashMap<String, Double> fixedPercepts);;
- public abstract double getProposition(ArrayList<Experiment> experimentsList, Experiment experiment);
+ /**
+ * Gets the proposition with the lowest value possible
+ * @return
+ */
+ public double getMinProposition();
- public abstract String getCoefsFormula();
-
- public abstract void updateModelWithExperiments(ArrayList<Experiment> experimentsList);
- public abstract void updateModelWithExperimentAndWeight(Experiment newExperiment, double weight, int numberOfPointsForRegression);
- public abstract void updateModel(Experiment newExperiment, double weight, int numberOfPointsForRegression);
- public abstract String coefsToString();
- public abstract double distance(Experiment experiment);
- public abstract ArrayList<Experiment> getFirstExperiments();
- public abstract void setFirstExperiments( ArrayList<Experiment> frstExp);
- public abstract boolean finishedFirstExperiments();
-
- public abstract Double[] getCoef();
- public abstract void setCoef(Double[] coef);
+ /**
+ * Gets the formula of the model
+ * @return
+ */
+ public String getCoefsFormula();
+
+ /**
+ * Update the model with a new experiment.
+ * @param newExperiment
+ * @param weight the weight of the new experiment in the compute of the model
+ */
+ public void updateModel(Experiment newExperiment, double weight);
+
+ public String coefsToString();
+
+ /**
+ * The distance between an experiment and the model.
+ * @param experiment
+ * @return
+ */
+ public double distance(Experiment experiment);
+
+ /**
+ * Gets the experiments used to properly initialize the model.
+ * @return
+ */
+ public ArrayList<Experiment> getFirstExperiments();
+
+ /**
+ * Sets the experiments used to properly initialize the model.
+ * This may not trigger an update of the model.
+ * @param frstExp
+ */
+ public void setFirstExperiments( ArrayList<Experiment> frstExp);
+
+ /**
+ * Tells if the model has enough experiments to produce a good prediction.
+ * For example, a regression need a number of experiments equals or superior to the number of dimension.
+ * @return
+ */
+ public boolean finishedFirstExperiments();
+
+ /**
+ * Gets coefficients of the model
+ * @return
+ */
+ public Double[] getCoef();
+
+ /**
+ * Sets coefficients of the model
+ * @return
+ */
+ public void setCoef(Double[] coef);
- public abstract TypeLocalModel getType();
+ /**
+ * Gets the {@link TypeLocalModel} corresponding to this LocalModel
+ */
+ public TypeLocalModel getType();
}
diff --git a/AMOEBAonAMAK/src/agents/context/localModel/LocalModelMillerRegression.java b/AMOEBAonAMAK/src/agents/context/localModel/LocalModelMillerRegression.java
index 87998fe51fe7f0fbd06036fa3e26bf74c662f8b9..1a9bcee4aa2ff7bff3472e4d74067a55d44af682 100644
--- a/AMOEBAonAMAK/src/agents/context/localModel/LocalModelMillerRegression.java
+++ b/AMOEBAonAMAK/src/agents/context/localModel/LocalModelMillerRegression.java
@@ -2,6 +2,7 @@ package agents.context.localModel;
import java.util.ArrayList;
import java.util.Arrays;
+import java.util.HashMap;
import java.util.List;
import agents.context.Context;
@@ -14,7 +15,9 @@ import utils.TRACE_LEVEL;
/**
* The Class LocalModelMillerRegression.
*/
-public class LocalModelMillerRegression extends LocalModel{
+public class LocalModelMillerRegression implements LocalModel{
+
+ private Context context;
/** The n parameters. */
private int nParameters;
@@ -34,7 +37,7 @@ public class LocalModelMillerRegression extends LocalModel{
* @param world the world
*/
public LocalModelMillerRegression(Context associatedContext) {
- super(associatedContext);
+ context = associatedContext;
ArrayList<Percept> var = associatedContext.getAmas().getPercepts();
this.nParameters = var.size();
regression = new Regression(nParameters,true);
@@ -42,7 +45,7 @@ public class LocalModelMillerRegression extends LocalModel{
}
public LocalModelMillerRegression(Context associatedContext, Double[] coefsCopy, List<Experiment> fstExperiments) {
- super(associatedContext);
+ context = associatedContext;
ArrayList<Percept> var = associatedContext.getAmas().getPercepts();
this.nParameters = var.size();
regression = new Regression(nParameters,true);
@@ -50,6 +53,16 @@ public class LocalModelMillerRegression extends LocalModel{
firstExperiments = new ArrayList<Experiment>(fstExperiments);
}
+ @Override
+ public void setContext(Context context) {
+ this.context = context;
+ }
+
+ @Override
+ public Context getContext() {
+ return context;
+ }
+
/**
* Sets the coef.
*
@@ -70,7 +83,8 @@ public class LocalModelMillerRegression extends LocalModel{
return coefs;
}
- public double getProposition(Context context) {
+ @Override
+ public double getProposition() {
ArrayList<Percept> percepts = context.getAmas().getPercepts();
@@ -92,11 +106,8 @@ public class LocalModelMillerRegression extends LocalModel{
- public double getProposition(Context context, double[] situation) {
-
- ArrayList<Percept> percepts = context.getAmas().getPercepts();
+ private double getProposition(Context context, double[] situation) {
-
double result = coefs[0];
if (coefs[0] == Double.NaN) System.exit(0);
@@ -111,28 +122,64 @@ public class LocalModelMillerRegression extends LocalModel{
return result;
}
- public double getMaxProposition(Context context) {
+ @Override
+ public double getMaxProposition() {
ArrayList<Percept> percepts = context.getAmas().getPercepts();
double result = coefs[0];
- if (coefs[0] == Double.NaN) System.exit(0);
+ if (coefs[0] == Double.NaN)
+ throw new ArithmeticException("First coeficient of model cannot be NaN");
for (int i = 1 ; i < coefs.length ; i++) {
-
- if (Double.isNaN(coefs[i])) coefs[i] = 0.0;
- if(coefs[i]>0) {
- result += coefs[i] * context.getRanges().get(percepts.get(i-1)).getEnd();
+ double coef = coefs[i];
+ if (Double.isNaN(coef)) coef = 0.0;
+ if(coef>0) {
+ result += coef * context.getRanges().get(percepts.get(i-1)).getEnd();
}
else {
- result += coefs[i] * context.getRanges().get(percepts.get(i-1)).getStart();
+ result += coef * context.getRanges().get(percepts.get(i-1)).getStart();
}
}
return result;
}
- public double getMinProposition(Context context) {
+ @Override
+ public HashMap<String, Double> getMaxWithConstraint(HashMap<String, Double> fixedPercepts){
+ ArrayList<Percept> percepts = context.getAmas().getPercepts();
+
+ HashMap<String, Double> result = new HashMap<String, Double>();
+ result.put("oracle", coefs[0]);
+
+ if (coefs[0] == Double.NaN)
+ throw new ArithmeticException("First coeficient of model cannot be NaN");
+
+ for (int i = 1 ; i < coefs.length ; i++) {
+ double coef = coefs[i];
+ if (Double.isNaN(coef)) coef = 0.0;
+ double pos;
+ Percept p = percepts.get(i-1);
+ if(fixedPercepts.containsKey(p.getName())) {
+ pos = fixedPercepts.get(p.getName());
+ } else {
+ if(coef>0) {
+ pos = context.getRanges().get(p).getEnd();
+ }
+ else {
+ pos = context.getRanges().get(p).getStart();
+ }
+ }
+ double value = coef * pos;
+ result.put("oracle", result.get("oracle") + value);
+ result.put(p.getName(), pos);
+ }
+
+ return result;
+ }
+
+ @Override
+ public double getMinProposition() {
ArrayList<Percept> percepts = context.getAmas().getPercepts();
double result = coefs[0];
@@ -153,31 +200,6 @@ public class LocalModelMillerRegression extends LocalModel{
return result;
}
- public double getProposition(ArrayList<Experiment> experimentsList, Experiment experiment) {
-
-
-
- if (experimentsList.size() == 1) {
- return experimentsList.get(0).getOracleProposition();
- }
- else {
- double result = coefs[0];
-
- if (coefs[0] == Double.NaN) System.exit(0);
-
- for (int i = 1 ; i < coefs.length ; i++) {
-
- if (Double.isNaN(coefs[i])) coefs[i] = 0.0;
- result += coefs[i] * experiment.getValuesAsArray()[i-1];
-
- }
-
- return result;
- }
-
-
- }
-
public double getProposition(Experiment experiment) {
if (coefs[0] == Double.NaN) System.exit(0);
@@ -194,6 +216,7 @@ public class LocalModelMillerRegression extends LocalModel{
}
+ @Override
public String getCoefsFormula() {
String result = "" +coefs[0];
// //System.out.println("Result 0" + " : " + result);
@@ -209,47 +232,8 @@ public class LocalModelMillerRegression extends LocalModel{
return result;
}
-
@Override
- public void updateModelWithExperiments(ArrayList<Experiment> experimentsList) {
-
- regression = new Regression(nParameters,true);
-
- for (Experiment exp : experimentsList) {
-
- regression.addObservation(exp.getValuesAsArray(), exp.getOracleProposition());
-
-
- }
-
-
-
- while (regression.getN() < experimentsList.get(0).getValuesAsLinkedHashMap().size() + 2) { //TODO : to improve
-
- regression.addObservation(experimentsList.get(0).getValuesAsArray(), experimentsList.get(0).getOracleProposition());
-
- System.out.println("Observations " + regression.getN());
-
- System.out.println(experimentsList.get(0).getValuesAsLinkedHashMap().toString());
- for (int i = 0 ; i < experimentsList.get(0).getValuesAsArray().length ; i++ ) {
- System.out.print(experimentsList.get(0).getValuesAsArray()[i] + " " );
- }
- System.out.println(experimentsList.get(0).getOracleProposition() + " " );
- }
-
-
-
- double[] coef = regression.regress().getParameterEstimates();
- coefs = new Double[coef.length];
- for(int i = 0; i < coef.length; i++) {
- coefs[i] = coef[i];
- }
-
-
- }
-
-
- public void updateModel(Experiment newExperiment, double weight, int numberOfPointsForRegression) {
+ public void updateModel(Experiment newExperiment, double weight) {
context.getAmas().getEnvironment().trace(TRACE_LEVEL.INFORM, new ArrayList<String>(Arrays.asList(context.getName(),"NEW POINT REGRESSION", "FIRST POINTS :", ""+firstExperiments.size(), "OLD MODEL :", coefsToString())));
if(firstExperiments.size()< (nParameters + 2)) {
@@ -257,7 +241,7 @@ public class LocalModelMillerRegression extends LocalModel{
updateModel();
}else {
- updateModelWithExperimentAndWeight(newExperiment, weight, numberOfPointsForRegression);
+ updateModelWithExperimentAndWeight(newExperiment, weight, context.getAmas().data.numberOfPointsForRegression);
}
context.getAmas().addSpatiallyAlteredContextForUnityUI(context);
@@ -460,12 +444,8 @@ public class LocalModelMillerRegression extends LocalModel{
return new Pair<double[][], double[]>(artificalExperiments, artificalResults);
}
-
-
-
-
-
+ @Override
public double distance(Experiment experiment) {
if (coefs[0] == Double.NaN) System.exit(0);
@@ -491,14 +471,17 @@ public class LocalModelMillerRegression extends LocalModel{
}
+ @Override
public ArrayList<Experiment> getFirstExperiments() {
return firstExperiments;
}
+ @Override
public void setFirstExperiments( ArrayList<Experiment> frstExp) {
firstExperiments = frstExp;
}
+ @Override
public String coefsToString() {
String coefsString = "";
if(coefs != null) {
@@ -509,6 +492,7 @@ public class LocalModelMillerRegression extends LocalModel{
return coefsString;
}
+ @Override
public boolean finishedFirstExperiments() {
return firstExperiments.size()>= (nParameters + 2);
}
diff --git a/AMOEBAonAMAK/src/agents/head/Head.java b/AMOEBAonAMAK/src/agents/head/Head.java
index 90995cdbe2aa62bfcc54dcd72ec4fcb03c0825f7..e0713dc3ca984abc162d7d3d0b3df1b5f0890669 100644
--- a/AMOEBAonAMAK/src/agents/head/Head.java
+++ b/AMOEBAonAMAK/src/agents/head/Head.java
@@ -17,9 +17,8 @@ import agents.percept.Percept;
import kernel.AMOEBA;
import ncs.NCS;
import utils.Pair;
-import utils.PickRandom;
import utils.PrintOnce;
-import utils.Quadruplet;
+import utils.RandomUtils;
import utils.TRACE_LEVEL;
/**
@@ -32,7 +31,6 @@ public class Head extends AmoebaAgent {
private Context bestContext;
private Context lastUsedContext;
private Context newContext;
- private String functionSelected;
HashMap<Percept, Double> currentSituation = new HashMap<Percept, Double>();
@@ -40,20 +38,8 @@ public class Head extends AmoebaAgent {
public Criticalities endogenousCriticalities;
private ArrayList<Context> activatedContexts = new ArrayList<Context>();
- private ArrayList<Context> activatedContextsCopyForUpdates = new ArrayList<Context>();
private ArrayList<Context> activatedNeighborsContexts = new ArrayList<Context>();
- private HashMap<Percept, ArrayList<Context>> partiallyActivatedContextInNeighbors = new HashMap<Percept, ArrayList<Context>>();
- private HashMap<Percept, ArrayList<Context>> partiallyActivatedContexts = new HashMap<Percept, ArrayList<Context>>();
- private HashMap<Percept, ArrayList<Context>> partialNeighborContexts = new HashMap<Percept, ArrayList<Context>>();
-
- private ArrayList<Context> contextsNeighborsByInfluence = new ArrayList<Context>();
-
- private HashMap<Percept, ContextPair<Context, Context>> requestSurroundings = new HashMap<Percept, ContextPair<Context, Context>>();
- private HashMap<Percept, ContextPair<Context, Context>> sharedIncompetenceContextPairs = new HashMap<Percept, ContextPair<Context, Context>>();
-
- private ArrayList<Context> contextsInCompetition = new ArrayList<Context>();
-
Queue<EndogenousRequest> endogenousRequests = new PriorityQueue<EndogenousRequest>(new Comparator<EndogenousRequest>(){
public int compare(EndogenousRequest r1, EndogenousRequest r2) {
return r2.getPriority().compareTo(r1.getPriority());
@@ -81,29 +67,6 @@ public class Head extends AmoebaAgent {
for(int i =0 ; i< 20;i++) {
getAmas().data.executionTimesSums[i]=0.0;
}
-
-
-
-
-
- }
-
-
-
- public void addPartiallyActivatedContext(Percept nonValidPercept, Context validContextExecptOnTheNonValidPercept) {
- partiallyActivatedContexts.get(nonValidPercept).add(validContextExecptOnTheNonValidPercept);
-
- }
-
- public void addPartiallyActivatedContextInNeighbors(Percept nonValidPercept,
- Context validContextExecptOnTheNonValidPercept) {
- partiallyActivatedContextInNeighbors.get(nonValidPercept).add(validContextExecptOnTheNonValidPercept);
-
- }
-
- public void addPartialRequestNeighborContext(Percept nonValidPercept, Context validContextNeighborExecptOnTheNonValidPercept) {
- partialNeighborContexts.get(nonValidPercept).add(validContextNeighborExecptOnTheNonValidPercept);
-
}
/**
@@ -280,28 +243,6 @@ public class Head extends AmoebaAgent {
}
- public double getSpatialGeneralizationScore() {
- return getAmas().data.spatialGeneralizationScore;
- }
-
- private boolean nearestLocalNeighbor(Context ctxt1, Context ctxt2) {
-
- boolean nearestLocalNeighborTest = false;
-
- for (Percept pct : getAmas().getPercepts()) {
-
- if (partiallyActivatedContextInNeighbors.get(pct).contains(ctxt1)
- && partiallyActivatedContextInNeighbors.get(pct).contains(ctxt2)) {
- nearestLocalNeighborTest = nearestLocalNeighborTest
- || (Math.abs(partiallyActivatedContextInNeighbors.get(pct).indexOf(ctxt1)
- - partiallyActivatedContextInNeighbors.get(pct).indexOf(ctxt2)) == 1);
- }
-
- }
-
- return nearestLocalNeighborTest;
- }
-
public double getMinMaxVolume() {
double minMaxVolume = 1;
for (Percept pct : getAmas().getPercepts()) {
@@ -336,7 +277,6 @@ public class Head extends AmoebaAgent {
} else {
getAmas().data.noBestContext = true;
- ArrayList<Context> allContexts = getAmas().getContexts();
Context nearestContext = this.getNearestContext(activatedNeighborsContexts);
if(nearestContext != null) {
getAmas().data.prediction = nearestContext.getActionProposal();
@@ -346,7 +286,7 @@ public class Head extends AmoebaAgent {
// to limit performance impact, we limit our search on a random sample.
// a better way would be to increase neighborhood.
PrintOnce.print("Play without oracle : no nearest context in neighbors, searching in a random sample. (only shown once)");
- List<Context> searchList = PickRandom.pickNRandomElements(getAmas().getContexts(), 100);
+ List<Context> searchList = RandomUtils.pickNRandomElements(getAmas().getContexts(), 100);
nearestContext = this.getNearestContext(searchList);
if(nearestContext != null) {
getAmas().data.prediction = nearestContext.getActionProposal();
@@ -386,12 +326,6 @@ public class Head extends AmoebaAgent {
getAmas().data.endogenousPredictionActivatedContextsSharedIncompetence = null;
getAmas().data.endogenousPredictionNContextsByInfluence = null;
- for (Percept pcpt : getAmas().getPercepts()) {
- requestSurroundings.get(pcpt).clear();
- sharedIncompetenceContextPairs.get(pcpt).clear();
- }
- contextsInCompetition.clear();
-
if (uniqueActivatedContext()) {
getAmas().data.endogenousPredictionActivatedContextsOverlaps = activatedContexts.get(0).getActionProposal();
getAmas().data.endogenousPredictionActivatedContextsOverlapsWorstDimInfluence = activatedContexts.get(0)
@@ -442,6 +376,7 @@ public class Head extends AmoebaAgent {
getAmas().data.maxConfidence = Double.NEGATIVE_INFINITY;
getAmas().data.minConfidence = Double.POSITIVE_INFINITY;
+ ArrayList<Context> contextsNeighborsByInfluence = new ArrayList<>();
for (Context ctxt : getAmas().getContexts()) {
if (ctxt.getConfidence() > getAmas().data.maxConfidence) {
@@ -516,20 +451,16 @@ public class Head extends AmoebaAgent {
// Test if there are surrounding contexts
boolean testSurroudingContext = false;
- for (Percept pct : getAmas().getPercepts()) {
-
- computeNearestContextsByPercept(pct);
-
- }
+ HashMap<Percept, ContextPair<Context, Context>> sharedIncompetenceContextPairs = getSharedIncompetenceContextPair();
// displayContexts();
for (Percept pcpt : getAmas().getPercepts()) {
- requestSurroundings.get(pcpt).print(pcpt);
if (sharedIncompetenceContextPairs.get(pcpt) != null) {
if (sharedIncompetenceContextPairs.get(pcpt).containTwoContexts()) {
testSurroudingContext = true;
+ break;
}
}
@@ -538,43 +469,40 @@ public class Head extends AmoebaAgent {
return testSurroudingContext;
}
- private void diplayListRanges(ArrayList<Context> list, Percept prct, String range) {
-
- System.out.print(range + " ranges list" + " ");
- for (Context ctxt : list) {
- System.out.print(ctxt.getRanges().get(prct).getRange(range) + " ");
+ private HashMap<Percept, ContextPair<Context, Context>> getSharedIncompetenceContextPair() {
+ HashMap<Percept, ContextPair<Context, Context>> sharedIncompetenceContextPairs = new HashMap<>();
+
+ for (Percept pct : getAmas().getPercepts()) {
+ ContextPair<Context, Context> nearestContexts = computeNearestContextsByPercept(pct);
+ if (nearestContexts.getL() != null && nearestContexts.getR() != null) {
+ sharedIncompetenceContextPairs.put(pct, nearestContexts);
+ }
}
- System.out.println(" ");
-
+
+ return sharedIncompetenceContextPairs;
}
- private void computeNearestContextsByPercept(Percept pct) {
- ContextPair<Context, Context> nearestContexts = new ContextPair(null, null);
+ private ContextPair<Context, Context> computeNearestContextsByPercept(Percept pct) {
+ ContextPair<Context, Context> nearestContexts = new ContextPair<>(null, null);
boolean startNeighbor = false;
boolean endNeighbor = false;
ArrayList<Context> activatedContextInOtherPercepts = getAllActivatedContextsExeptForOnePercept(pct);
-
if (activatedContextInOtherPercepts.size() > 0) {
-
-
CustomComparator rangeStartComparator = new CustomComparator(pct, "start");
Collections.sort(activatedContextInOtherPercepts, rangeStartComparator);
- // diplayListRanges(activatedContextInOtherPercepts, pct, "start");
for (Context ctxt : activatedContextInOtherPercepts) {
if (ctxt.getRanges().get(pct).getRange("start") > pct.getValue() && !startNeighbor) {
nearestContexts.setR(ctxt);
startNeighbor = true;
}
-
}
CustomComparator rangeEndComparator = new CustomComparator(pct, "end");
Collections.sort(activatedContextInOtherPercepts, rangeEndComparator);
Collections.reverse(activatedContextInOtherPercepts);
- // diplayListRanges(activatedContextInOtherPercepts, pct, "end");
for (Context ctxt : activatedContextInOtherPercepts) {
if (ctxt.getRanges().get(pct).getRange("end") < pct.getValue() && !endNeighbor) {
@@ -583,65 +511,26 @@ public class Head extends AmoebaAgent {
}
}
- // nearestContexts.print(pct);
- requestSurroundings.put(pct, nearestContexts);
-
- if (nearestContexts.getL() != null && nearestContexts.getR() != null) {
- sharedIncompetenceContextPairs.put(pct, nearestContexts);
- }
- } else {
-
- }
-
- }
-
- private ArrayList<Context> getAllActivatedContextsExeptForOnePercept(Percept onePercept) {
-
- return partiallyActivatedContexts.get(onePercept);
- }
-
- private ArrayList<Context> getAllActivatedNeighborContextsExeptForOnePercept(Percept onePercept) {
-
- return partialNeighborContexts.get(onePercept);
- }
-
- public void displayPartiallyActivatedContexts() {
- //////// System.out.println("PARTIALLY ACTIVATED CONTEXTS");
- for (Percept pct : partiallyActivatedContexts.keySet()) {
- //// System.out.print(pct.getName() + " : ");
- if (partiallyActivatedContexts.get(pct).size() > 0)
- for (Context ctxt : partiallyActivatedContexts.get(pct)) {
- //// System.out.print(ctxt.getName() + " ; ");
- }
- //////// System.out.println(" ");
- }
- }
-
- public void displayContexts() {
- //////// System.out.println("CONTEXTS");
- for (Context ctxt : getAmas().getContexts()) {
- //// System.out.print(ctxt.getName() + " ; ");
- }
- //////// System.out.println(" ");
- }
-
- private Percept getDifferentPercept(Percept p) {
- for (Percept pct : partiallyActivatedContexts.keySet()) {
- if (p != pct) {
- return pct;
- }
}
- return null;
+ return nearestContexts;
}
- private boolean contextActivateInOtherPerceptsThan(Context ctxt, Percept p1, Percept p2) {
- boolean test = true;
- for (Percept prct : partiallyActivatedContexts.keySet()) {
- if (prct != p1 && prct != p2) {
- test = test & partiallyActivatedContexts.get(prct).contains(ctxt);
+ /**
+ * Return a list of contexts that have been activated by all percepts except a particular one.
+ * @param pct
+ * @return
+ */
+ private ArrayList<Context> getAllActivatedContextsExeptForOnePercept(Percept pct) {
+ ArrayList<Percept> percepts = new ArrayList<>(getAmas().getPercepts());
+ percepts.remove(pct);
+ ArrayList<Context> ret = new ArrayList<>(getAmas().getContexts());
+ ret.removeIf(c -> {
+ for(Percept p : percepts) {
+ if(!p.activateContext(c)) return true;
}
- }
- return test;
+ return false;
+ });
+ return ret;
}
private void NCS_EndogenousCompetition() {
@@ -746,6 +635,7 @@ public class Head extends AmoebaAgent {
double smallestDistanceBetweenContexts = Double.POSITIVE_INFINITY;
double currentDistance;
+ HashMap<Percept, ContextPair<Context, Context>> sharedIncompetenceContextPairs = getSharedIncompetenceContextPair();
for (Percept pct : sharedIncompetenceContextPairs.keySet()) {
if (sharedIncompetenceContextPairs.get(pct) != null) {
if (sharedIncompetenceContextPairs.get(pct).containTwoContexts()) {
@@ -826,11 +716,6 @@ public class Head extends AmoebaAgent {
}
- private void NCS_EndogenousIncompetence() {
- // Extrapolation of contexts by creating twin contexts that will give the
- // prediction
- }
-
private void NCSDetection_Create_New_Context() {
/* Finally, head agent check the need for a new context agent */
@@ -977,27 +862,16 @@ public class Head extends AmoebaAgent {
getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("------------------------------------------------------------------------------------"
+ "---------------------------------------- NCS DETECTION POTENTIAL REQUESTS")));
-
-
-
if (activatedNeighborsContexts.size() > 1) {
-
int i = 1;
for (Context ctxt : activatedNeighborsContexts) {
-
for (Context otherCtxt : activatedNeighborsContexts.subList(i, activatedNeighborsContexts.size())) {
-
-
if(!this.isDying() && !ctxt.isDying()) {
EndogenousRequest potentialRequest = ctxt.endogenousRequest(otherCtxt);
-
if(potentialRequest != null) {
addEndogenousRequest(potentialRequest);
-
}
}
-
-
}
i++;
}
@@ -1005,100 +879,17 @@ public class Head extends AmoebaAgent {
getEnvironment().trace(TRACE_LEVEL.STATE, new ArrayList<String>(Arrays.asList("ENDO REQUESTS", ""+endogenousRequests.size())));
for(EndogenousRequest endoRequest : endogenousRequests) {
-
getEnvironment().trace(TRACE_LEVEL.STATE, new ArrayList<String>(Arrays.asList("" + endoRequest)));
-
}
}
- private void selfAnalysationOfContexts2() {
-
- if (activatedContexts.size() > 1) {
- selfAnalysationOfContextOnSeveralActivatedContexts();
-
- } else if (activatedContexts.size() == 1) {
-
-// double distanceToOracleForActivatedContext = activatedContexts.get(0).getLocalModel().distance(activatedContexts.get(0).getCurrentExperiment());
-// //System.out.println(distanceToOracleForActivatedContext + " ******************************************************************DISTANCE TO MODEL : " );
-// if(activatedNeighborsContexts.size()>1) {
-//
-// Context closestContextToOracle = null;
-// double minDistanceToOraclePredictionInNeighbors = Double.POSITIVE_INFINITY;
-// double currentDistanceToOraclePrediction = 0.0;
-// for(Context contextNeighbor : activatedNeighborsContexts) {
-// if(contextNeighbor != activatedContexts.get(0)) {
-// currentDistanceToOraclePrediction = contextNeighbor.getLocalModel().distance(contextNeighbor.getCurrentExperiment());
-// if(currentDistanceToOraclePrediction<minDistanceToOraclePredictionInNeighbors) {
-// minDistanceToOraclePredictionInNeighbors = currentDistanceToOraclePrediction;
-// closestContextToOracle = contextNeighbor;
-// }
-// }
-// }
-//
-// if(minDistanceToOraclePredictionInNeighbors>distanceToOracleForActivatedContext) {
-// //System.out.println("OLD COEFS " + activatedContexts.get(0).getLocalModel().coefsToString());
-// activatedContexts.get(0).getLocalModel().updateModel(activatedContexts.get(0).getCurrentExperiment(),learningSpeed,numberOfPointsForRegression);
-// //System.out.println("NEW COEFS " + activatedContexts.get(0).getLocalModel().coefsToString());
-//
-// }else {
-// closestContextToOracle.solveNCS_IncompetentHead(this);
-// //activatedContexts.get(0).setLocalModel(new LocalModelMillerRegression(world, activatedContexts.get(0), closestContextToOracle.getLocalModel().getCoef(),closestContextToOracle.getLocalModel().getFirstExperiments()));
-// //LocalModelAgent remplacementModel = new LocalModelMillerRegression(world, activatedContexts.get(0), closestContextToOracle.getLocalModel().getCoef().clone());
-// //activatedContexts.get(0).setLocalModel(remplacementModel);
-//
-// }
-//
-//
-// }else {
-// //System.out.println("OLD COEFS " + activatedContexts.get(0).getLocalModel().coefsToString());
-// activatedContexts.get(0).getLocalModel().updateModel(activatedContexts.get(0).getCurrentExperiment(),learningSpeed,numberOfPointsForRegression);
-// //System.out.println("NEW COEFS " + activatedContexts.get(0).getLocalModel().coefsToString());
-//
-//
-// }
-
- selfAnalysationOfContextOnUniqueActivatedContext();
-
- }
-
- for (Context ctxt : activatedNeighborsContexts) {
-
- if (!activatedContexts.contains(ctxt)) {
- ctxt.NCSDetection_Uselessness();
- }
-
- }
- }
-
- private void selfAnalysationOfContexts3() {
-
- if (activatedContexts.size() > 1) {
-
- selfAnalysationOfContextOnSeveralActivatedContexts();
-
- } else if (activatedContexts.size() == 1) {
-
- selfAnalysationOfContextOnUniqueActivatedContext();
-
- }
-
- for (Context ctxt : activatedNeighborsContexts) {
-
- if (!activatedContexts.contains(ctxt)) {
- ctxt.NCSDetection_Uselessness();
- }
-
- }
- }
-
private void selfAnalysationOfContexts4() {
getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("------------------------------------------------------------------------------------"
+ "---------------------------------------- SELF ANALYSIS OF CTXT")));
double currentDistanceToOraclePrediction;
- Context closestContextToOracle = null;
double minDistanceToOraclePrediction = Double.POSITIVE_INFINITY;
for (Context activatedContext : activatedContexts) {
@@ -1110,21 +901,18 @@ public class Head extends AmoebaAgent {
getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE", activatedContext.getName(),
"" + activatedContext.getLocalModel().distance(activatedContext.getCurrentExperiment()))));
if (!activatedContext.getLocalModel().finishedFirstExperiments()) {
- activatedContext.getLocalModel().updateModel(activatedContext.getCurrentExperiment(), getAmas().data.learningSpeed,
- getAmas().data.numberOfPointsForRegression);
+ activatedContext.getLocalModel().updateModel(activatedContext.getCurrentExperiment(), getAmas().data.learningSpeed);
getAmas().data.contextNotFinished = true;
}
else if (currentDistanceToOraclePrediction < getAverageRegressionPerformanceIndicator()) {
//else if (currentDistanceToOraclePrediction < regressionPerformance.getPerformanceIndicator()) {
- activatedContext.getLocalModel().updateModel(activatedContext.getCurrentExperiment(), getAmas().data.learningSpeed,
- getAmas().data.numberOfPointsForRegression);
+ activatedContext.getLocalModel().updateModel(activatedContext.getCurrentExperiment(), getAmas().data.learningSpeed);
}
if (currentDistanceToOraclePrediction < minDistanceToOraclePrediction) {
minDistanceToOraclePrediction = currentDistanceToOraclePrediction;
- closestContextToOracle = activatedContext;
}
if (!getAmas().data.contextNotFinished) {
@@ -1164,136 +952,6 @@ public class Head extends AmoebaAgent {
}
}
- private void selfAnalysationOfContextOnUniqueActivatedContext() {
-
- if (activatedNeighborsContexts.size() > 0) {
- double distanceToOracleForActivatedContext = activatedContexts.get(0).getLocalModel()
- .distance(activatedContexts.get(0).getCurrentExperiment());
- Quadruplet<Context, Double, Context, Double> closestAndFarestContextsToPredictionWithDistance = closestAndFarestContextsToPrediction();
-
- double distanceToMin = Math
- .abs(distanceToOracleForActivatedContext - closestAndFarestContextsToPredictionWithDistance.getB());
- double distanceToMax = Math
- .abs(distanceToOracleForActivatedContext - closestAndFarestContextsToPredictionWithDistance.getD());
-
- if (distanceToMin < distanceToMax) {
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE",
- activatedContexts.get(0).getName(), "" + distanceToOracleForActivatedContext)));
- activatedContexts.get(0).getLocalModel().updateModel(activatedContexts.get(0).getCurrentExperiment(),
- getAmas().data.learningSpeed, getAmas().data.numberOfPointsForRegression);
-
- }
- } else {
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(
- Arrays.asList("MODEL DISTANCE", activatedContexts.get(0).getName(), "" + activatedContexts.get(0)
- .getLocalModel().distance(activatedContexts.get(0).getCurrentExperiment()))));
- activatedContexts.get(0).getLocalModel().updateModel(activatedContexts.get(0).getCurrentExperiment(),
- getAmas().data.learningSpeed, getAmas().data.numberOfPointsForRegression);
- }
-
- // world.trace(new ArrayList<String>(Arrays.asList("MODEL
- // DISTANCE",activatedContexts.get(0).getName(),
- // ""+activatedContexts.get(0).getLocalModel().distance(activatedContexts.get(0).getCurrentExperiment()))));
- // activatedContexts.get(0).getLocalModel().updateModel(activatedContexts.get(0).getCurrentExperiment(),learningSpeed,numberOfPointsForRegression);
- activatedContexts.get(0).analyzeResults3(this, activatedContexts.get(0));
- }
-
- private Pair<Context, Double> maxModelDistanceToOraclePredictionInNeighbors() {
- Context farestContextToOracle = null;
- double maxDistanceToOraclePredictionInNeighbors = Double.NEGATIVE_INFINITY;
- double currentDistanceToOraclePrediction = 0.0;
- for (Context contextNeighbor : activatedNeighborsContexts) {
- if (contextNeighbor != activatedContexts.get(0)) {
- currentDistanceToOraclePrediction = contextNeighbor.getLocalModel()
- .distance(contextNeighbor.getCurrentExperiment());
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE", contextNeighbor.getName(),
- "" + currentDistanceToOraclePrediction)));
- if (currentDistanceToOraclePrediction > maxDistanceToOraclePredictionInNeighbors) {
- maxDistanceToOraclePredictionInNeighbors = currentDistanceToOraclePrediction;
- farestContextToOracle = contextNeighbor;
- }
- }
- }
- return new Pair<Context, Double>(farestContextToOracle, maxDistanceToOraclePredictionInNeighbors);
- }
-
- private Pair<Context, Double> minModelDistanceToOraclePredictionInNeighbors() {
- Context closestContextToOracle = null;
- double minDistanceToOraclePredictionInNeighbors = Double.POSITIVE_INFINITY;
- double currentDistanceToOraclePrediction = 0.0;
- for (Context contextNeighbor : activatedNeighborsContexts) {
- if (contextNeighbor != activatedContexts.get(0)) {
- currentDistanceToOraclePrediction = contextNeighbor.getLocalModel()
- .distance(contextNeighbor.getCurrentExperiment());
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE", contextNeighbor.getName(),
- "" + currentDistanceToOraclePrediction)));
- if (currentDistanceToOraclePrediction < minDistanceToOraclePredictionInNeighbors) {
- minDistanceToOraclePredictionInNeighbors = currentDistanceToOraclePrediction;
- closestContextToOracle = contextNeighbor;
- }
- }
- }
- return new Pair<Context, Double>(closestContextToOracle, minDistanceToOraclePredictionInNeighbors);
- }
-
- private Quadruplet<Context, Double, Context, Double> closestAndFarestContextsToPrediction() {
- Context farestContextToOracle = null;
- double maxDistanceToOraclePredictionInNeighbors = Double.NEGATIVE_INFINITY;
- Context closestContextToOracle = null;
- double minDistanceToOraclePredictionInNeighbors = Double.POSITIVE_INFINITY;
-
- double currentDistanceToOraclePrediction = 0.0;
- for (Context contextNeighbor : activatedNeighborsContexts) {
- if (contextNeighbor != activatedContexts.get(0)) {
- currentDistanceToOraclePrediction = contextNeighbor.getLocalModel()
- .distance(contextNeighbor.getCurrentExperiment());
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE", contextNeighbor.getName(),
- "" + currentDistanceToOraclePrediction)));
- if (currentDistanceToOraclePrediction > maxDistanceToOraclePredictionInNeighbors) {
- maxDistanceToOraclePredictionInNeighbors = currentDistanceToOraclePrediction;
- farestContextToOracle = contextNeighbor;
- }
- if (currentDistanceToOraclePrediction < minDistanceToOraclePredictionInNeighbors) {
- minDistanceToOraclePredictionInNeighbors = currentDistanceToOraclePrediction;
- closestContextToOracle = contextNeighbor;
- }
- }
- }
- return new Quadruplet<Context, Double, Context, Double>(closestContextToOracle,
- minDistanceToOraclePredictionInNeighbors, farestContextToOracle,
- maxDistanceToOraclePredictionInNeighbors);
- }
-
- private void selfAnalysationOfContextOnSeveralActivatedContexts() {
-
- Context closestContextToOracle = null;
- double minDistanceToOraclePrediction = Double.POSITIVE_INFINITY;
- double currentDistanceToOraclePrediction = 0.0;
-
- for (Context activatedContext : activatedContexts) {
- currentDistanceToOraclePrediction = activatedContext.getLocalModel()
- .distance(activatedContext.getCurrentExperiment());
-
- getEnvironment().trace(TRACE_LEVEL.DEBUG, new ArrayList<String>(Arrays.asList("MODEL DISTANCE", activatedContext.getName(),
- "" + activatedContext.getLocalModel().distance(activatedContext.getCurrentExperiment()))));
- if (currentDistanceToOraclePrediction < minDistanceToOraclePrediction) {
- minDistanceToOraclePrediction = currentDistanceToOraclePrediction;
- closestContextToOracle = activatedContext;
- }
-
- }
-
- closestContextToOracle.getLocalModel().updateModel(closestContextToOracle.getCurrentExperiment(), getAmas().data.learningSpeed,
- getAmas().data.numberOfPointsForRegression);
-
- activatedContextsCopyForUpdates = new ArrayList<Context>(activatedContexts);
- for (Context activatedContext : activatedContexts) {
- activatedContext.analyzeResults3(this, closestContextToOracle);
-
- }
- activatedContexts = activatedContextsCopyForUpdates;
- }
-
private void setNearestContextAsBestContext() {
Context nearestContext = this.getNearestContext(activatedNeighborsContexts);
@@ -1306,25 +964,6 @@ public class Head extends AmoebaAgent {
bestContext = nearestContext;
}
- private void setNearestGoodContextAsBestContext() {
- Context nearestContext = this.getNearestGoodContext(activatedNeighborsContexts);
-
- if (nearestContext != null) {
- getAmas().data.prediction = nearestContext.getActionProposal();
- } else {
- getAmas().data.prediction = 0.0;
- }
-
- bestContext = nearestContext;
- }
-
- /**
- * Endogenous feedback.
- */
- private void endogenousFeedback() {
- // bestContext.growRanges(this);
- }
-
/**
* Gets the nearest good context.
*
@@ -1364,28 +1003,6 @@ public class Head extends AmoebaAgent {
}
- /**
- * Gets the distance to nearest good context.
- *
- * @param allContext the all context
- * @return the distance to nearest good context
- */
- private Pair<Context, Double> getNearestGoodContextWithDistance(ArrayList<Context> contextNeighbors) {
- double d = Double.MAX_VALUE;
- Context nearestGoodContext = null;
- for (Context c : contextNeighbors) {
- if (Math.abs((c.getActionProposal() - getAmas().data.oracleValue)) <= getAmas().data.predictionPerformance.getPerformanceIndicator()
- && c != newContext && !c.isDying()) {
- if (getExternalDistanceToContext(c) < d) {
- d = getExternalDistanceToContext(c);
- nearestGoodContext = c;
- }
- }
- }
- return new Pair<Context, Double>(nearestGoodContext, d);
-
- }
-
private Pair<Context, Double> getbestContextInNeighborsWithDistanceToModel(ArrayList<Context> contextNeighbors) {
double d = Double.MAX_VALUE;
Context bestContextInNeighbors = null;
@@ -1464,19 +1081,6 @@ public class Head extends AmoebaAgent {
return d;
}
- private double getDistanceToContext(Context context) {
- double totalDistance = 0.0;
- for (Percept p : getAmas().getPercepts()) {
- double distance = context.distance(p, p.getValue());
- if (distance > 0) {
- totalDistance += distance;
- }
-
- }
-
- return totalDistance;
- }
-
/**
* One of proposed context was good.
*
@@ -1951,24 +1555,6 @@ public class Head extends AmoebaAgent {
getAmas().data.useOracle = !getAmas().data.useOracle;
}
- /**
- * Gets the function selected.
- *
- * @return the function selected
- */
- public String getFunctionSelected() {
- return functionSelected;
- }
-
- /**
- * Sets the function selected.
- *
- * @param functionSelected the new function selected
- */
- public void setFunctionSelected(String functionSelected) {
- this.functionSelected = functionSelected;
- }
-
/**
* Gets the n propositions received.
*
@@ -2076,86 +1662,6 @@ public class Head extends AmoebaAgent {
this.getAmas().data.prediction = prediction;
}
- public ArrayList<Context> getPartiallyActivatedContexts(Percept pct) {
- return partiallyActivatedContexts.get(pct);
- }
-
- public ArrayList<Context> getPartiallyActivatedNeighborContexts(Percept pct) {
- return partialNeighborContexts.get(pct);
- }
-
- public HashMap<Percept, ContextPair<Context, Context>> getRequestSurroundings() {
- return requestSurroundings;
- }
-
- public boolean requestSurroundingContains(Context ctxt) {
-
- for (Percept pct : requestSurroundings.keySet()) {
-
- if (requestSurroundings.get(pct).contains(ctxt)) {
- return true;
- }
- }
- return false;
-
- }
-
- public ArrayList<Context> getContextsInCompetition() {
- return contextsInCompetition;
- }
-
- public Head clone() throws CloneNotSupportedException {
- return (Head) super.clone();
- }
-
- public void addRequestNeighbor(Context ctxt) {
- if (!activatedNeighborsContexts.contains(ctxt)) {
-
- activatedNeighborsContexts.add(ctxt);
- }
- }
-
- public void addActivatedContext(Context ctxt) {
- if (!activatedContexts.contains(ctxt)) {
- activatedContexts.add(ctxt);
-
-
- for (Context cxt : activatedContexts) {
-
- }
- }
- }
-
- public void addActivatedContextCopy(Context ctxt) {
- if (!activatedContextsCopyForUpdates.contains(ctxt)) {
- activatedContextsCopyForUpdates.add(ctxt);
-
-
- for (Context cxt : activatedContextsCopyForUpdates) {
-
- }
- }
- }
-
- public void removeActivatedContext(Context ctxt) {
- if (activatedContexts.contains(ctxt)) {
- activatedContexts.remove(ctxt);
- }
- }
-
- public void removeActivatedContextCopy(Context ctxt) {
- if (activatedContextsCopyForUpdates.contains(ctxt)) {
- activatedContextsCopyForUpdates.remove(ctxt);
- }
- }
-
- public void removeRequestNeighbor(Context ctxt) {
- if (activatedNeighborsContexts.contains(ctxt)) {
-
- activatedNeighborsContexts.remove(ctxt);
- }
- }
-
public ArrayList<Context> getActivatedNeighborsContexts() {
return activatedNeighborsContexts;
}
@@ -2164,34 +1670,9 @@ public class Head extends AmoebaAgent {
activatedNeighborsContexts = neighbors;
}
-
- public ArrayList<Context> getContextNeighborsByInfluence() {
- return contextsNeighborsByInfluence;
- }
-
- public void displayActivatedNeighborsContexts() {
- for (Context ctxt : activatedNeighborsContexts) {
-
- }
- }
-
- public void clearActivatedNeighborsContexts() {
- activatedNeighborsContexts.clear();
- }
-
- public void clearContextdNeighborsByInfluence() {
- contextsNeighborsByInfluence.clear();
- }
-
public void clearAllUseableContextLists() {
activatedContexts.clear();
activatedNeighborsContexts.clear();
- contextsNeighborsByInfluence.clear();
- for (Percept pct : getAmas().getPercepts()) {
- partiallyActivatedContexts.get(pct).clear();
- partiallyActivatedContextInNeighbors.get(pct).clear();
- partialNeighborContexts.get(pct).clear();
- }
}
public Double getMaxRadiusForContextCreation(Percept pct) {
@@ -2372,21 +1853,6 @@ public class Head extends AmoebaAgent {
}
- public void updatePartiallyActivatedNeighbors() {
-
- for(Percept pct: getAmas().getPercepts()) {
-
- pct.computeContextNeighborsValidity(activatedNeighborsContexts);
-
- }
-
- for(Context ctxt : activatedNeighborsContexts) {
-
- ctxt.computeContextNeighborsValidity();
-
- }
- }
-
public boolean isVoid(HashMap<Percept, Double> request) {
boolean test;
@@ -2410,18 +1876,11 @@ public class Head extends AmoebaAgent {
int numberOfRegressions = 0;
if(activatedNeighborsContexts.size()>0) {
double meanRegressionPerformanceIndicator = 0.0;
-
for(Context ctxt : activatedNeighborsContexts) {
-
-
- if(ctxt.regressionPerformance != null) {
meanRegressionPerformanceIndicator += ctxt.regressionPerformance.performanceIndicator;
numberOfRegressions+=1;
- }
-
-
}
-
+ assert numberOfRegressions != 0;
return meanRegressionPerformanceIndicator/numberOfRegressions;
}
else{
@@ -2429,16 +1888,8 @@ public class Head extends AmoebaAgent {
}
}
-
-
-
-
- // -----------------
- // AMOEBAonAMAK ---
- // -----------------
public void proposition(Context c) {
-
activatedContexts.add(c);
}
@@ -2448,19 +1899,8 @@ public class Head extends AmoebaAgent {
getAmas().data.maxConfidence = Double.NEGATIVE_INFINITY;
getAmas().data.minConfidence = Double.POSITIVE_INFINITY;
- // getAmas().data.mappingPerformance.setPerformanceIndicator(world.getMappingErrorAllowed());//
- // Math.pow(world.getMappingErrorAllowed(),
- // world.getScheduler().getPercepts().size());
-
criticalities = new Criticalities(getAmas().data.numberOfCriticityValuesForAverage);
endogenousCriticalities = new Criticalities(getAmas().data.numberOfCriticityValuesForAverageforVizualisation);
}
-
- public void addPercept(Percept pct) {
- partiallyActivatedContextInNeighbors.put(pct, new ArrayList<Context>());
- partiallyActivatedContexts.put(pct, new ArrayList<Context>());
- partialNeighborContexts.put(pct, new ArrayList<Context>());
- requestSurroundings.put(pct, new ContextPair<Context, Context>(null, null));
- sharedIncompetenceContextPairs.put(pct, new ContextPair<Context, Context>(null, null));
- }
+
}
diff --git a/AMOEBAonAMAK/src/agents/percept/Percept.java b/AMOEBAonAMAK/src/agents/percept/Percept.java
index 064969461e53732961f008123fc2106b850b5bfe..52c9a398ded9453bac8fb6692c977517f61b5050 100644
--- a/AMOEBAonAMAK/src/agents/percept/Percept.java
+++ b/AMOEBAonAMAK/src/agents/percept/Percept.java
@@ -53,8 +53,6 @@ public class Percept extends AmoebaAgent {
value = getAmas().getPerceptions(this.name);
ajustMinMax();
computeContextProjectionValidityOptimized();
- //computeContextProjectionValidity();
-
}
public void computeContextProjectionValidity() {
@@ -124,7 +122,7 @@ public class Percept extends AmoebaAgent {
}
for (Context c : contexts) {
- if (contextProjections.get(c).contains(this.value)) {
+ if (activateContext(c)) {
validContextProjection.add(c);
}
}
@@ -137,7 +135,7 @@ public class Percept extends AmoebaAgent {
}
for (Context c : neighborsContexts) {
- if(contextProjections.get(c).inNeighborhood()) {
+ if(inNeighborhood(c)) {
neighborContextProjection.add(c);
}
}
@@ -145,6 +143,24 @@ public class Percept extends AmoebaAgent {
logger().debug("CYCLE "+getAmas().getCycle(), "%s's valid contexts : %s", toString(), validContextProjection.toString());
}
+
+ /**
+ * Return true if the context is activated by this percept.
+ * @param context
+ * @return
+ */
+ public boolean activateContext(Context context) {
+ return contextProjections.get(context).contains(this.value);
+ }
+
+ /**
+ * Return true if the context is in the neighborhood of this percept.
+ * @param context
+ * @return
+ */
+ public boolean inNeighborhood(Context context) {
+ return contextProjections.get(context).inNeighborhood();
+ }
/**
diff --git a/AMOEBAonAMAK/src/experiments/F_XY_System.java b/AMOEBAonAMAK/src/experiments/F_XY_System.java
index 86a0d739f772a9e98e7c03d1aa7b10eaeb764f28..30f3fdfdf8fa9706fea2abee71c0a9307b06fc2d 100644
--- a/AMOEBAonAMAK/src/experiments/F_XY_System.java
+++ b/AMOEBAonAMAK/src/experiments/F_XY_System.java
@@ -48,7 +48,7 @@ public class F_XY_System implements StudiedSystem {
out.put("px0", x);
out.put("px1", y);
- out.put("oracle", result);
+ out.put("oracle", 0.0);
return out;
}
diff --git a/AMOEBAonAMAK/src/experiments/MinimalMain.java b/AMOEBAonAMAK/src/experiments/MinimalMain.java
index 60218985a5ed9718f003c4cfe7f8cf21f96a65a2..b9b0bde35821831dfe658b762a50be32ceb0540a 100644
--- a/AMOEBAonAMAK/src/experiments/MinimalMain.java
+++ b/AMOEBAonAMAK/src/experiments/MinimalMain.java
@@ -1,12 +1,7 @@
package experiments;
-import java.util.HashMap;
-
-import fr.irit.smac.amak.Configuration;
-import fr.irit.smac.amak.tools.Log;
import kernel.AMOEBA;
import kernel.StudiedSystem;
-import kernel.backup.SaveHelperDummy;
/**
* The most minimal main possible producing a functioning amoeba.
@@ -16,61 +11,12 @@ import kernel.backup.SaveHelperDummy;
public class MinimalMain {
public static void main(String[] args) throws InterruptedException {
- Configuration.commandLineMode = true;
- Log.defaultMinLevel = Log.Level.INFORM;
// create a system to be studied
- StudiedSystem studiedSystem = new NDimCube(50.0, 3);
+ StudiedSystem studiedSystem = new F_XY_System(50.0);
// create the amoeba
// Make sure the path to the config file is correct.
- AMOEBA amoeba = new AMOEBA("resources/threeDimensionsLauncher.xml", studiedSystem);
- amoeba.saver = new SaveHelperDummy();
+ AMOEBA amoeba = new AMOEBA("resources/twoDimensionsLauncher.xml", studiedSystem);
// a window should have appeared, allowing you to control and visualize the amoeba.
-
- Thread.sleep(5000);
-
- long start = System.currentTimeMillis();
- long end = System.currentTimeMillis();
- for(int i = 0; i < 10000; i++) {
- studiedSystem.playOneStep();
- amoeba.learn(studiedSystem.getOutput());
- if(i%100 == 99) {
- end = System.currentTimeMillis();
- System.out.println("Time for 100 learn: "+(end-start)/1000.0);
- start = System.currentTimeMillis();
- }
- }
-
- start = System.currentTimeMillis();
- end = System.currentTimeMillis();
- for(int i = 0; i < 1000; i++) {
- studiedSystem.playOneStep();
- amoeba.request(studiedSystem.getOutput());
- if(i%100 == 99) {
- end = System.currentTimeMillis();
- System.out.println("Time for 100 request: "+(end-start)/1000.0);
- start = System.currentTimeMillis();
- }
- }
-
- start = System.currentTimeMillis();
- end = System.currentTimeMillis();
- for(int i = 0; i < 1000; i++) {
- studiedSystem.playOneStep();
- HashMap<String, Double> req = studiedSystem.getOutput();
- req.remove("px1");
- req.remove("px2");
- req.remove("px3");
- req.remove("oracle");
- amoeba.maximize(req);
- if(i%100 == 99) {
- end = System.currentTimeMillis();
- System.out.println("Time for 100 maximize: "+(end-start)/1000.0);
- start = System.currentTimeMillis();
- }
- }
-
- System.exit(0);
-
}
}
diff --git a/AMOEBAonAMAK/src/experiments/SimpleReinforcement.java b/AMOEBAonAMAK/src/experiments/SimpleReinforcement.java
new file mode 100644
index 0000000000000000000000000000000000000000..b63df76532b63de09980521021efc64b13d007eb
--- /dev/null
+++ b/AMOEBAonAMAK/src/experiments/SimpleReinforcement.java
@@ -0,0 +1,125 @@
+package experiments;
+
+import java.io.File;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.Random;
+
+import fr.irit.smac.amak.Configuration;
+import fr.irit.smac.amak.ui.drawables.Drawable;
+import fr.irit.smac.amak.ui.drawables.DrawableOval;
+import gui.AmoebaWindow;
+import javafx.scene.paint.Color;
+import kernel.AMOEBA;
+import utils.Pair;
+import utils.RandomUtils;
+import utils.XmlConfigGenerator;
+
+/**
+ * Train an amoeba on a simple reinforcement task.
+ * The goal of the task is to get to the center. When the position of the agent cross 0, it gets a reward of 100.
+ * The agent can only moves in 2 directions, of a distance of 1. Moving give a reward of -1.
+ * If the agent moves outside of the allowed range, it gets a reward of -100.
+ * @author Hugo
+ *
+ */
+public class SimpleReinforcement {
+
+ private Random rand = new Random();
+ private double x = 0;
+ private double reward = 0;
+ private Drawable pos;
+
+ public static void main(String[] args) {
+ ArrayList<Pair<String, Boolean>> sensors = new ArrayList<>();
+ sensors.add(new Pair<String, Boolean>("p1", false));
+ sensors.add(new Pair<String, Boolean>("a1", true));
+ File config;
+ try {
+ config = File.createTempFile("config", "xml");
+ XmlConfigGenerator.makeXML(config, sensors);
+ } catch (IOException e) {
+ e.printStackTrace();
+ System.exit(1);
+ return; // now compilator know config is initialized
+ }
+
+ AMOEBA amoeba = new AMOEBA(config.getAbsolutePath(), null);
+ SimpleReinforcement env = new SimpleReinforcement();
+
+ Random r = new Random();
+ HashMap<String, Double> state = env.reset();
+ HashMap<String, Double> state2;
+ for(int i = 0; i < 10000; i++) {
+ state.remove("oracle");
+ HashMap<String, Double> action = amoeba.maximize(state);
+ if(r.nextDouble() < 0.2 || action.get("oracle").equals(Double.NEGATIVE_INFINITY) ) {
+ System.out.println("Random action");
+ action.put("a1", (r.nextBoolean() ? 10.0 : -10.0));
+ }
+ state2 = env.step(action.get("a1"));
+ action.put("oracle", state2.get("oracle"));
+ action.put("p1", state.get("p1"));
+ System.out.println(action);
+ amoeba.learn(action);
+ state = state2;
+ }
+ }
+
+ /**
+ * Must be called AFTER an AMOEBA with GUI
+ */
+ public SimpleReinforcement() {
+ Configuration.commandLineMode = false;
+ AmoebaWindow instance = AmoebaWindow.instance();
+ pos = new DrawableOval(0.5, 0.5, 1, 1);
+ pos.setColor(new Color(0.5, 0.0, 0.0, 0.5));
+ instance.mainVUI.add(pos);
+ instance.mainVUI.createAndAddRectangle(-50, -0.25, 100, 0.5);
+ instance.mainVUI.createAndAddRectangle(-0.25, -1, 0.5, 2);
+
+
+
+ }
+
+ public HashMap<String, Double> step(double action){
+ if(action == 0.0) action = rand.nextDouble();
+ if(action > 0.0) action = Math.ceil(action);
+ if(action < 0.0 ) action = Math.floor(action);
+ if(action > 1.0) action = 1.0;
+ if(action < -1.0) action = -1.0;
+ double oldX = x;
+ x = x + action;
+ if(x < -50.0 || x > 50.0) {
+ x = RandomUtils.nextDouble(rand, -50.0, Math.nextUp(50.0));
+ reward = -100.0;
+ } else if(x == 0.0 || sign(oldX) != sign(x)) {
+ // win !
+ reward = 100.0;
+ x = RandomUtils.nextDouble(rand, -50.0, Math.nextUp(50.0));
+ } else {
+ reward = -1.0;
+ }
+ HashMap<String, Double> ret = new HashMap<>();
+ ret.put("p1", x);
+ ret.put("oracle", reward);
+ pos.move(x+0.5, 0.5);
+ return ret;
+ }
+
+ public HashMap<String, Double> reset(){
+ x = RandomUtils.nextDouble(rand, -50.0, Math.nextUp(50.0));
+ reward = 0.0;
+
+ HashMap<String, Double> ret = new HashMap<>();
+ ret.put("p1", x);
+ ret.put("oracle", reward);
+ return ret;
+ }
+
+ private int sign(double x) {
+ return x < 0 ? -1 : 1;
+ }
+
+}
diff --git a/AMOEBAonAMAK/src/experiments/TestingMain.java b/AMOEBAonAMAK/src/experiments/TestingMain.java
index 20332e2005b0b34e1b3965b4469a2a1362c2ad39..55bfd61be6a785e91185c2cf2eff2c396e810a15 100644
--- a/AMOEBAonAMAK/src/experiments/TestingMain.java
+++ b/AMOEBAonAMAK/src/experiments/TestingMain.java
@@ -16,7 +16,7 @@ import kernel.backup.SaveHelperDummy;
public class TestingMain {
public static void main(String[] args) throws InterruptedException {
- Configuration.commandLineMode = true;
+ Configuration.commandLineMode = false;
Log.defaultMinLevel = Log.Level.INFORM;
// create a system to be studied
StudiedSystem studiedSystem = new NDimCube(50.0, 3);
@@ -30,7 +30,7 @@ public class TestingMain {
long start = System.currentTimeMillis();
long end = System.currentTimeMillis();
- for(int i = 0; i < 10000; i++) {
+ for(int i = 0; i < 1000; i++) {
studiedSystem.playOneStep();
amoeba.learn(studiedSystem.getOutput());
if(i%100 == 99) {
diff --git a/AMOEBAonAMAK/src/experiments/XmlConfigGenerator.java b/AMOEBAonAMAK/src/experiments/XmlConfigGenerator.java
deleted file mode 100644
index 00f5086a85ab7d5363024529ae1cf8a115749274..0000000000000000000000000000000000000000
--- a/AMOEBAonAMAK/src/experiments/XmlConfigGenerator.java
+++ /dev/null
@@ -1,69 +0,0 @@
-package experiments;
-
-import java.io.File;
-import java.io.FileNotFoundException;
-import java.io.FileWriter;
-import java.io.IOException;
-
-public class XmlConfigGenerator {
-
- public XmlConfigGenerator() {
-
- }
-
- public static String makeLineXML(String sensorName, String source) {
- return "\t\t<Sensor Name=\""+sensorName+"\" Source=\""+source+"\"></Sensor>\n";
- }
-
- public static void makeXML(String XMLFile, int dimension) {
- String path = "Ressources";
-
- try {
- File file = new File(path+"/"+XMLFile);
- FileWriter fw = new FileWriter(file);
-
- //Make content
- String str = "";
- str += "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n" +
- "<System>\n" +
- "\n" +
- "\t<!-- General config options -->\n" +
- "\t<Configuration>\n" +
- "\t\t<Learning allowed = \"true\" creationOfNewContext = \"true\" loadPresetContext = \"false\"></Learning> \n" +
- "\t</Configuration>\n" +
- "\n" +
- "\t<StartingAgents>\n";
-
- for(int j = 0; j < dimension; ++j) {
- str += makeLineXML("px"+j,"x"+j);
- }
-
- str += "\t\t<Controller Name=\"Controller\" Oracle=\"test\"></Controller>\n\n";
-
-
-
- str += "\t</StartingAgents>\n" +
- "\n" +
- "</System>";
-
- //Write and close file
- fw.write(str);
- fw.close();
- } catch (FileNotFoundException e) {
- e.printStackTrace();
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
-
- public static void main(String[] args) {
- // TODO Auto-generated method stub
-
-
- String XMLConfigFile = "nDimensionLauncher.xml";
-
- XmlConfigGenerator.makeXML(XMLConfigFile, 2);
-
-
- }
-}
diff --git a/AMOEBAonAMAK/src/experiments/nDimensionsLaunchers/F_N_Launcher.java b/AMOEBAonAMAK/src/experiments/nDimensionsLaunchers/F_N_Launcher.java
index 5a9e38b3e7007b38ae8e0c90a8bf84420d1a82eb..2e0310717c904b7ffee6afa26c57f1f89c95cb61 100644
--- a/AMOEBAonAMAK/src/experiments/nDimensionsLaunchers/F_N_Launcher.java
+++ b/AMOEBAonAMAK/src/experiments/nDimensionsLaunchers/F_N_Launcher.java
@@ -4,13 +4,8 @@ import java.io.File;
import java.io.IOException;
import java.io.Serializable;
import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.HashMap;
-import agents.context.Context;
import experiments.FILE;
-import experiments.F_XY_System;
-import experiments.XmlConfigGenerator;
import fr.irit.smac.amak.Configuration;
import gui.AmoebaWindow;
import kernel.AMOEBA;
@@ -90,6 +85,14 @@ public class F_N_Launcher implements Serializable {
// }
// long end = System.currentTimeMillis();
// System.out.println("Done in : " + (end - start) );
+//
+// start = System.currentTimeMillis();
+// for (int i = 0; i < nbCycle; ++i) {
+// studiedSystem.playOneStep();
+// amoeba.request(studiedSystem.getOutput());
+// }
+// end = System.currentTimeMillis();
+// System.out.println("Done in : " + (end - start) );
// /* XP PIERRE */
diff --git a/AMOEBAonAMAK/src/gui/ContextRendererFX.java b/AMOEBAonAMAK/src/gui/ContextRendererFX.java
index cfe62c32fd7d5c9909ad1710e570cf216962df7f..75ddf4125607f21b3ef19ca0dcaaea006105be4d 100644
--- a/AMOEBAonAMAK/src/gui/ContextRendererFX.java
+++ b/AMOEBAonAMAK/src/gui/ContextRendererFX.java
@@ -83,7 +83,7 @@ public class ContextRendererFX extends RenderStrategy {
*/
public DrawableRectangle getDrawable() {
if (!context.isDying() && drawable == null) {
- drawable = new DrawableContext(0, 0, 10, 10, context);
+ drawable = new DrawableContext(0, 0, 0, 0, context);
AmoebaWindow.instance().mainVUI.add(drawable);
}
return drawable;
diff --git a/AMOEBAonAMAK/src/kernel/AMOEBA.java b/AMOEBAonAMAK/src/kernel/AMOEBA.java
index d05843c62e72bac2fd2c9abb45de07fb06c2625c..502a41fd835f1a65aaeb864a7585496afe3a1d58 100644
--- a/AMOEBAonAMAK/src/kernel/AMOEBA.java
+++ b/AMOEBAonAMAK/src/kernel/AMOEBA.java
@@ -429,78 +429,23 @@ public class AMOEBA extends Amas<World> implements IAMOEBA {
if(good) pac.add(c);
}
- ArrayList<Pair<HashMap<String, Double>, Double>> sol = new ArrayList<>();
+ ArrayList<HashMap<String, Double>> sol = new ArrayList<>();
for(Context c : pac) {
- sol.add(maximiseContext(known, percepts, unknown, c));
+ sol.add(c.getLocalModel().getMaxWithConstraint(known));
}
HashMap<String, Double> max = new HashMap<>();
- // set default value if no solution
- for(Percept p : unknown) {
- max.put(p.getName(), 0.0);
- }
+
Double maxValue = Double.NEGATIVE_INFINITY;
+ max.put("oracle", maxValue);
//find best solution
- for(Pair<HashMap<String, Double>, Double> s : sol) {
- if(s.getB() > maxValue) {
- maxValue = s.getB();
- max = s.getA();
+ for(HashMap<String, Double> s : sol) {
+ if(s.get("oracle") > maxValue) {
+ maxValue = s.get("oracle");
+ max = s;
}
}
- max.put("oracle", maxValue);
return max;
}
-
- private Pair<HashMap<String, Double>, Double> maximiseDummy(HashMap<String, Double> known,
- ArrayList<Percept> percepts, ArrayList<Percept> unknown, Context c) {
- HashMap<String, Double> res = new HashMap<>();
- for(Percept p : unknown) {
- res.put(p.getName(), c.getRangeByPercept(p).getCenter());
- }
- HashMap<String, Double> tmpReq = new HashMap<>(res);
- HashMap<String, Double> old = perceptions;
- perceptions = tmpReq;
- Pair<HashMap<String, Double>, Double> ret = new Pair<>(res, c.getActionProposal());
- perceptions = old;
- return ret;
- }
-
- //TODO tests !
- private Pair<HashMap<String, Double>, Double> maximiseContext(HashMap<String, Double> known,
- ArrayList<Percept> percepts, ArrayList<Percept> unknown, Context c) {
- HashMap<String, Double> res = new HashMap<>();
-
- Double[] coefs = c.getLocalModel().getCoef();
- double[] vCoefs = new double[coefs.length-1];
- for(int i = 1; i < coefs.length; i++) {
- vCoefs[i-1] = coefs[i];
- }
- LinearObjectiveFunction fct = new LinearObjectiveFunction(vCoefs, coefs[0]);
- ArrayList<LinearConstraint> constraints = new ArrayList<>();
- //TODO : problem : we are not sure that the order of percepts is the same as coefs
- int i = 0;
- for(String p : known.keySet()) {
- double[] cf = new double[percepts.size()];
- cf[i++] = 1.0;
- constraints.add(new LinearConstraint(cf, Relationship.EQ, known.get(p)));
- }
- int unknowStart = i;
- for(Percept p : unknown) {
- double[] cf = new double[percepts.size()];
- cf[i++] = 1.0;
- constraints.add(new LinearConstraint(cf, Relationship.GEQ, c.getRangeByPercept(p).getStart()));
- constraints.add(new LinearConstraint(cf, Relationship.LEQ, c.getRangeByPercept(p).getEnd()));
- }
- SimplexSolver solver = new SimplexSolver();
- LinearConstraintSet set = new LinearConstraintSet(constraints);
- PointValuePair sol = solver.optimize(fct, set, GoalType.MAXIMIZE);
- for(Percept p : unknown) {
- //TODO check if the order match
- res.put(p.getName(), sol.getFirst()[unknowStart++]);
- }
-
- Pair<HashMap<String, Double>, Double> ret = new Pair<>(res, sol.getSecond());
- return ret;
- }
public LocalModel buildLocalModel(Context context) {
switch (localModel) {
@@ -734,15 +679,6 @@ public class AMOEBA extends Amas<World> implements IAMOEBA {
alteredContexts.add(context);
}
- /**
- * Gets the altered contexts.
- *
- * @return the altered contexts
- */
- public ArrayList<Context> getAlteredContexts() {
- return alteredContexts;
- }
-
/**
* Return the current set of valid contexts.
*
@@ -832,7 +768,6 @@ public class AMOEBA extends Amas<World> implements IAMOEBA {
public void addPercept(Percept pct) {
percepts = null;
- head.addPercept(pct);
}
}
diff --git a/AMOEBAonAMAK/src/utils/PickRandom.java b/AMOEBAonAMAK/src/utils/RandomUtils.java
similarity index 63%
rename from AMOEBAonAMAK/src/utils/PickRandom.java
rename to AMOEBAonAMAK/src/utils/RandomUtils.java
index ae94493c4a00b50e6654ab8722eac7f98de66117..9440f17725430ccad2abcee4edb6470d61072134 100644
--- a/AMOEBAonAMAK/src/utils/PickRandom.java
+++ b/AMOEBAonAMAK/src/utils/RandomUtils.java
@@ -5,7 +5,7 @@ import java.util.List;
import java.util.Random;
import java.util.concurrent.ThreadLocalRandom;
-public class PickRandom {
+public class RandomUtils {
/**
* Pick N random element from the list. if n is bigger than the list, return the list.
@@ -40,4 +40,19 @@ public class PickRandom {
public static <E> List<E> pickNRandomElements(List<E> list, int n) {
return pickNRandomElements(list, n, ThreadLocalRandom.current());
}
+
+ /**
+ * Generate a pseudorandom double values, conforming to the given origin (inclusive) and bound(exclusive).
+ * @param rand
+ * @param origin the origin (inclusive) of the random value
+ * @param bound the bound (exclusive) of the random value
+ * @return
+ */
+ public static double nextDouble(Random rand, double origin, double bound) {
+ double r = rand.nextDouble();
+ r = r * (bound - origin) + origin;
+ if (r >= bound) // correct for rounding
+ r = Math.nextDown(bound);
+ return r;
+ }
}
diff --git a/AMOEBAonAMAK/src/utils/XmlConfigGenerator.java b/AMOEBAonAMAK/src/utils/XmlConfigGenerator.java
new file mode 100644
index 0000000000000000000000000000000000000000..3819e2af7c35df94763eab320fbacd78491f335e
--- /dev/null
+++ b/AMOEBAonAMAK/src/utils/XmlConfigGenerator.java
@@ -0,0 +1,61 @@
+package utils;
+
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Locale;
+
+public class XmlConfigGenerator {
+
+ public static String header = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n";
+ public static String systemStart = "<System>\n";
+ public static String configurationStart = "<Configuration>\n";
+ public static String configuration = " <Learning allowed = \"%s\" creationOfNewContext = \"%s\" loadPresetContext = \"%s\" />\n";
+ public static String configurationEnd = "</Configuration>\n";
+ public static String agentsStart = "<StartingAgents>\n";
+ public static String sensor = " <Sensor Name=\"%s\" Enum=\"%s\" />\n";
+ public static String controller = " <Controller Name=\"Controller\">\n" +
+ " <ErrorMargin ErrorAllowed=\"%f\" AugmentationFactorError=\"%f\" DiminutionFactorError=\"%f\" MinErrorAllowed=\"%f\" NConflictBeforeAugmentation=\"%d\" NSuccessBeforeDiminution=\"%d\" />\n" +
+ " </Controller>\n";
+ public static String agentsEnd = "</StartingAgents>\n";
+ public static String systemEnd = "</System>\n";
+
+ private static String makeSensor(String sensorName, boolean isEnum) {
+ return String.format(sensor, sensorName, ""+isEnum);
+ }
+
+ private static String makeConfiguration(boolean isLearningAllowed, boolean isCreationOfNewContext, boolean isLoadContext) {
+ return String.format(configuration, ""+isLearningAllowed, ""+isCreationOfNewContext, ""+isLoadContext);
+ }
+
+ private static String makeController(double errorAllowed, double augmentationFactorError, double diminutionFactorError, double minErrorAllowed, int nConflictBeforeAugmentation, int nSuccessBeforeDiminution) {
+ return String.format(Locale.ENGLISH, controller, errorAllowed, augmentationFactorError, diminutionFactorError, minErrorAllowed, nConflictBeforeAugmentation, nSuccessBeforeDiminution);
+ }
+
+ public static void makeXML(File file, List<Pair<String, Boolean>> sensors) throws IOException {
+ try(FileWriter fw = new FileWriter(file)){
+ fw.write(header);
+ fw.write(systemStart);
+ fw.write(configurationStart);
+ fw.write(makeConfiguration(true, true, false));
+ fw.write(configurationEnd);
+ fw.write(agentsStart);
+ for(Pair<String, Boolean> s : sensors) {
+ fw.write(makeSensor(s.getA(), s.getB()));
+ }
+ fw.write(makeController(1.0, 0.5, 0.5, 0.01, 50, 50));
+ fw.write(agentsEnd);
+ fw.write(systemEnd);
+ }
+ }
+
+ public static void makeXML(File file, int dimension) throws IOException {
+ ArrayList<Pair<String, Boolean>> sensors = new ArrayList<>(dimension);
+ for(int i = 1; i <= dimension; i++) {
+ sensors.add(new Pair<String, Boolean>("p"+i, false));
+ }
+ makeXML(file, sensors);
+ }
+}