/**
* Apply the NNSSLGenerator method.
*
* @return
*/
public Pair<PrototypeSet, PrototypeSet> applyAlgorithm() {
System.out.print("\nThe algorithm SELF TRAINING is starting...\n Computing...\n");
PrototypeSet labeled;
PrototypeSet unlabeled;
labeled =
new PrototypeSet(
trainingDataSet.getAllDifferentFromClass(
this.numberOfClass)); // Selecting labeled prototypes from the training set.
unlabeled = new PrototypeSet(trainingDataSet.getFromClass(this.numberOfClass));
PrototypeSet tranductive = new PrototypeSet(this.transductiveDataSet.clone());
PrototypeSet test = new PrototypeSet(this.testDataSet.clone());
// We have to return the classification done.
for (int i = 0; i < this.transductiveDataSet.size(); i++) {
tranductive
.get(i)
.setFirstOutput((labeled.nearestTo(this.transductiveDataSet.get(i))).getOutput(0));
}
for (int i = 0; i < this.testDataSet.size(); i++) {
test.get(i).setFirstOutput((labeled.nearestTo(this.testDataSet.get(i))).getOutput(0));
}
// Transductive Accuracy
System.out.println(
"AccTrs ="
+ KNN.classficationAccuracy1NN(labeled, this.transductiveDataSet)
* 100.
/ this.transductiveDataSet.size());
// test accuracy
System.out.println(
"AccTst ="
+ KNN.classficationAccuracy1NN(labeled, this.testDataSet)
* 100.
/ this.testDataSet.size());
return new Pair<PrototypeSet, PrototypeSet>(tranductive, test);
}
/**
* General main for all the prototoype generators Arguments: 0: Filename with the training data
* set to be condensed. 1: Filename wich will contain the test data set
*
* @param args Arguments of the main function.
*/
public static void main(String[] args) {
Parameters.setUse("AVG", "");
Parameters.assertBasicArgs(args);
PrototypeSet training = PrototypeGenerationAlgorithm.readPrototypeSet(args[0]);
PrototypeSet test = PrototypeGenerationAlgorithm.readPrototypeSet(args[1]);
AVG generator = new AVG(training);
PrototypeSet resultingSet = generator.execute();
int accuracy1NN = KNN.classficationAccuracy(resultingSet, test);
generator.showResultsOfAccuracy(Parameters.getFileName(), accuracy1NN, test);
}
protected int NN(int nSel, double conj[][], double ejemplo[]) {
double mindist, dist;
int nneigh = -1;
mindist = Double.POSITIVE_INFINITY;
for (int i = 0; i < nSel; i++) {
dist = KNN.distancia(conj[i], ejemplo);
if (dist < mindist) {
mindist = dist;
nneigh = i;
}
}
return nneigh;
}
int nextNeighbour(boolean marcas[], double datos[][], int ej, Vector<Integer> vecinos) {
int i, j, k;
int pos = -1;
double distmin = Double.POSITIVE_INFINITY;
double distancia;
double centroid[];
double prototipo[];
/*Computation of the previous centroid*/
centroid = new double[datos[0].length];
prototipo = new double[datos[0].length];
for (j = 0; j < datos[0].length; j++) {
centroid[j] = 0;
for (k = 0; k < vecinos.size(); k++) {
centroid[j] += datos[vecinos.elementAt(k).intValue()][j];
}
}
for (i = 0; i < datos.length; i++) {
if (marcas[i] && i != ej) {
for (j = 0; j < datos[0].length; j++) {
prototipo[j] = centroid[j] + datos[i][j];
prototipo[j] /= (vecinos.size() + 1);
}
distancia = KNN.distancia(datos[ej], prototipo);
if (distancia < distmin) {
distmin = distancia;
pos = i;
}
}
}
return pos;
}
public void ejecutar() {
int i, j, l, m;
double alfai;
int nClases;
int claseObt;
boolean marcas[];
boolean notFound;
int init;
int clasSel[];
int baraje[];
int pos, tmp;
String instanciasIN[];
String instanciasOUT[];
long tiempo = System.currentTimeMillis();
/* Getting the number of differents classes */
nClases = 0;
for (i = 0; i < clasesTrain.length; i++) if (clasesTrain[i] > nClases) nClases = clasesTrain[i];
nClases++;
/* Shuffle the train set */
baraje = new int[datosTrain.length];
Randomize.setSeed(semilla);
for (i = 0; i < datosTrain.length; i++) baraje[i] = i;
for (i = 0; i < datosTrain.length; i++) {
pos = Randomize.Randint(i, datosTrain.length - 1);
tmp = baraje[i];
baraje[i] = baraje[pos];
baraje[pos] = tmp;
}
/*
* Inicialization of the flagged instaces vector for a posterior
* elimination
*/
marcas = new boolean[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) marcas[i] = false;
if (datosTrain.length > 0) {
// marcas[baraje[0]] = true; //the first instance is included always
nSel = n_p;
if (nSel < nClases) nSel = nClases;
} else {
System.err.println("Input dataset is empty");
nSel = 0;
}
clasSel = new int[nClases];
System.out.print("Selecting initial neurons... ");
// at least, there must be 1 neuron of each class at the beginning
init = nClases;
for (i = 0; i < nClases && i < datosTrain.length; i++) {
pos = Randomize.Randint(0, datosTrain.length - 1);
tmp = 0;
while ((clasesTrain[pos] != i || marcas[pos]) && tmp < datosTrain.length) {
pos = (pos + 1) % datosTrain.length;
tmp++;
}
if (tmp < datosTrain.length) marcas[pos] = true;
else init--;
// clasSel[i] = i;
}
for (i = init; i < Math.min(nSel, datosTrain.length); i++) {
tmp = 0;
pos = Randomize.Randint(0, datosTrain.length - 1);
while (marcas[pos]) {
pos = (pos + 1) % datosTrain.length;
tmp++;
}
// if(i<nClases){
// notFound = true;
// do{
// for(j=i-1;j>=0 && notFound;j--){
// if(clasSel[j] == clasesTrain[pos])
// notFound = false;
// }
// if(!notFound)
// pos = Randomize.Randint (0, datosTrain.length-1);
// }while(!notFound);
// }
// clasSel[i] = clasesTrain[pos];
marcas[pos] = true;
init++;
}
nSel = init;
System.out.println("Initial neurons selected: " + nSel);
/* Building of the S set from the flags */
conjS = new double[nSel][datosTrain[0].length];
clasesS = new int[nSel];
for (m = 0, l = 0; m < datosTrain.length; m++) {
if (marcas[m]) { // the instance must be copied to the solution
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[m][j];
}
clasesS[l] = clasesTrain[m];
l++;
}
}
alfai = alpha;
boolean change = true;
/* Body of the LVQ algorithm. */
// Train the network
for (int it = 0; it < T && change; it++) {
change = false;
alpha = alfai;
for (i = 1; i < datosTrain.length; i++) {
// search for the nearest neuron to training instance
pos = NN(nSel, conjS, datosTrain[baraje[i]]);
// nearest neuron labels correctly the class of training
// instance?
if (clasesS[pos] != clasesTrain[baraje[i]]) { // NO - repel
// the neuron
for (j = 0; j < conjS[pos].length; j++) {
conjS[pos][j] = conjS[pos][j] - alpha * (datosTrain[baraje[i]][j] - conjS[pos][j]);
}
change = true;
} else { // YES - migrate the neuron towards the input vector
for (j = 0; j < conjS[pos].length; j++) {
conjS[pos][j] = conjS[pos][j] + alpha * (datosTrain[baraje[i]][j] - conjS[pos][j]);
}
}
alpha = nu * alpha;
}
// Shuffle again the training partition
baraje = new int[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) baraje[i] = i;
for (i = 0; i < datosTrain.length; i++) {
pos = Randomize.Randint(i, datosTrain.length - 1);
tmp = baraje[i];
baraje[i] = baraje[pos];
baraje[pos] = tmp;
}
}
System.out.println(
"LVQ " + relation + " " + (double) (System.currentTimeMillis() - tiempo) / 1000.0 + "s");
// Classify the train data set
instanciasIN = new String[datosReferencia.length];
instanciasOUT = new String[datosReferencia.length];
for (i = 0; i < datosReferencia.length; i++) {
/* Classify the instance selected in this iteration */
Attribute a = Attributes.getOutputAttribute(0);
int tipo = a.getType();
claseObt = KNN.evaluacionKNN2(1, conjS, clasesS, datosReferencia[i], nClases);
if (tipo != Attribute.NOMINAL) {
instanciasIN[i] = new String(String.valueOf(clasesReferencia[i]));
instanciasOUT[i] = new String(String.valueOf(claseObt));
} else {
instanciasIN[i] = new String(a.getNominalValue(clasesReferencia[i]));
instanciasOUT[i] = new String(a.getNominalValue(claseObt));
}
}
escribeSalida(
ficheroSalida[0], instanciasIN, instanciasOUT, entradas, salida, nEntradas, relation);
// Classify the test data set
normalizarTest();
instanciasIN = new String[datosTest.length];
instanciasOUT = new String[datosTest.length];
for (i = 0; i < datosTest.length; i++) {
/* Classify the instance selected in this iteration */
Attribute a = Attributes.getOutputAttribute(0);
int tipo = a.getType();
claseObt = KNN.evaluacionKNN2(1, conjS, clasesS, datosTest[i], nClases);
if (tipo != Attribute.NOMINAL) {
instanciasIN[i] = new String(String.valueOf(clasesTest[i]));
instanciasOUT[i] = new String(String.valueOf(claseObt));
} else {
instanciasIN[i] = new String(a.getNominalValue(clasesTest[i]));
instanciasOUT[i] = new String(a.getNominalValue(claseObt));
}
}
escribeSalida(
ficheroSalida[1], instanciasIN, instanciasOUT, entradas, salida, nEntradas, relation);
// Print the network to a file
printNetworkToFile(ficheroSalida[2], referencia.getHeader());
}
public double crearMetodo(Instances pTrain, Instances pTest, Instances pEvaluar) {
// Cargamos las instancias clasificadas
Instances train = pTrain;
Instances test = pTest;
double bestAccuracy = 0;
double accuracy = 0;
KNN bestKNN = null;
int bestL = 0;
int indice = 0;
// Movemos el valor de L para cambiar la cantidad de las muestras
for (int i = 2; i < 16; i++) {
ArrayList<ArrayList<Instance>> L = crearArrays(i);
// Le metemos las instancias a cada ArrayList
for (ArrayList<Instance> grupoInstancias : L) {
grupoInstancias.addAll(rellenarConInstancias(train));
}
// Creamos los clasificadores
ArrayList<KNN> clasificadores = new ArrayList<KNN>(i);
for (int j = 0; j < i; j++) {
try {
clasificadores.set(j, new KNN(L.get(j)));
} catch (IndexOutOfBoundsException e) {
clasificadores.add(new KNN(L.get(j)));
}
}
// Array de resultados
ArrayList<Instance> Resultados = new ArrayList<Instance>();
// Clasificamos las instancias
for (int j = 0; j < test.numInstances(); j++) {
// Creo un Array de Instancias para guardar los resultados
ArrayList<Instance> clasificadas = new ArrayList<Instance>();
// Clasificamos la instancia con todo el grupo de KNN
for (KNN knn : clasificadores) {
clasificadas.add(knn.clasificarInstacia(2, 2, test.instance(j)));
}
// Escogemos la clase mayoritaria del Array de clasificadas.
Integer[] clases = new Integer[test.classAttribute().numValues()];
// Inicio el array a cero
for (int k = 0; k < clases.length; k++) {
clases[k] = 0;
}
// Asigno el numero de instancias con esa clase.
for (Instance instancia : clasificadas) {
clases[
instancia
.classAttribute()
.indexOfValue(instancia.stringValue(instancia.classAttribute()))] +=
1;
}
// Elegir la clase mayoritaria y introducir a resultado
int numero = 0; // numero
indice = 0; // indice
for (int k = 0; k < clases.length; k++) {
if (clases[k] > numero) {
numero = clases[k];
indice = k;
}
}
clasificadas.get(0).setClassValue(clasificadas.get(0).classAttribute().value(indice));
Resultados.add(clasificadas.get(0));
Evaluador ev = new Evaluador(pEvaluar);
accuracy = ev.EvaluateModel(Resultados);
if (accuracy > bestAccuracy) {
bestAccuracy = accuracy;
bestL = i;
}
}
}
System.out.println("Mejor L: " + bestL);
System.out.println("Mejor accuracy: " + bestAccuracy);
return accuracy;
}
/**
* The main method of the class that includes the operations of the algorithm. It includes all the
* operations that the algorithm has and finishes when it writes the output information into
* files.
*/
public void run() {
int S[];
int i, j, l, m;
int nPos = 0, nNeg = 0;
int posID;
int nClases;
int pos;
int baraje[];
int tmp;
double conjS[][];
int clasesS[];
int tamS = 0;
int claseObt;
int cont;
int busq;
boolean marcas[];
int nSel;
double conjS2[][];
int clasesS2[];
double minDist, dist;
long tiempo = System.currentTimeMillis();
/*CNN PART*/
/*Count of number of positive and negative examples*/
for (i = 0; i < clasesTrain.length; i++) {
if (clasesTrain[i] == 0) nPos++;
else nNeg++;
}
if (nPos > nNeg) {
tmp = nPos;
nPos = nNeg;
nNeg = tmp;
posID = 1;
} else {
posID = 0;
}
/*Inicialization of the candidates set*/
S = new int[datosTrain.length];
for (i = 0; i < S.length; i++) S[i] = Integer.MAX_VALUE;
/*Inserting an element of mayority class*/
Randomize.setSeed(semilla);
pos = Randomize.Randint(0, clasesTrain.length - 1);
while (clasesTrain[pos] == posID) pos = (pos + 1) % clasesTrain.length;
S[tamS] = pos;
tamS++;
/*Insert all subset of minority class*/
for (i = 0; i < clasesTrain.length; i++) {
if (clasesTrain[i] == posID) {
S[tamS] = i;
tamS++;
}
}
/*Algorithm body. We resort randomly the instances of T and compare with the rest of S.
If an instance doesn´t classified correctly, it is inserted in S*/
baraje = new int[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) baraje[i] = i;
for (i = 0; i < datosTrain.length; i++) {
pos = Randomize.Randint(i, clasesTrain.length - 1);
tmp = baraje[i];
baraje[i] = baraje[pos];
baraje[pos] = tmp;
}
for (i = 0; i < datosTrain.length; i++) {
if (clasesTrain[i] != posID) { // only for mayority class instances
/*Construction of the S set from the previous vector S*/
conjS = new double[tamS][datosTrain[0].length];
clasesS = new int[tamS];
for (j = 0; j < tamS; j++) {
for (l = 0; l < datosTrain[0].length; l++) conjS[j][l] = datosTrain[S[j]][l];
clasesS[j] = clasesTrain[S[j]];
}
/*Do KNN to the instance*/
claseObt = KNN.evaluacionKNN(k, conjS, clasesS, datosTrain[baraje[i]], 2);
if (claseObt != clasesTrain[baraje[i]]) { // fail in the class, it is included in S
Arrays.sort(S);
busq = Arrays.binarySearch(S, baraje[i]);
if (busq < 0) {
S[tamS] = baraje[i];
tamS++;
}
}
}
}
/*Construction of the S set from the previous vector S*/
conjS = new double[tamS][datosTrain[0].length];
clasesS = new int[tamS];
for (j = 0; j < tamS; j++) {
for (l = 0; l < datosTrain[0].length; l++) conjS[j][l] = datosTrain[S[j]][l];
clasesS[j] = clasesTrain[S[j]];
}
/*TOMEK LINKS PART*/
/*Inicialization of the instance flagged vector of the S set*/
marcas = new boolean[conjS.length];
for (i = 0; i < conjS.length; i++) {
marcas[i] = true;
}
nSel = conjS.length;
for (i = 0; i < conjS.length; i++) {
minDist = Double.POSITIVE_INFINITY;
pos = 0;
for (j = 0; j < conjS.length; j++) {
if (i != j) {
dist = KNN.distancia(conjS[i], conjS[j]);
if (dist < minDist) {
minDist = dist;
pos = j;
}
}
}
if (clasesS[i] != clasesS[pos]) {
if (clasesS[i] != posID) {
if (marcas[i] == true) {
marcas[i] = false;
nSel--;
}
} else {
if (marcas[pos] == true) {
marcas[pos] = false;
nSel--;
}
}
}
}
/*Construction of the S set from the flags*/
conjS2 = new double[nSel][conjS[0].length];
clasesS2 = new int[nSel];
for (m = 0, l = 0; m < conjS.length; m++) {
if (marcas[m]) { // the instance will evaluate
for (j = 0; j < conjS[0].length; j++) {
conjS2[l][j] = conjS[m][j];
}
clasesS2[l] = clasesS[m];
l++;
}
}
System.out.println(
"CNN_TomekLinks "
+ relation
+ " "
+ (double) (System.currentTimeMillis() - tiempo) / 1000.0
+ "s");
OutputIS.escribeSalida(
ficheroSalida[0], conjS2, clasesS2, entradas, salida, nEntradas, relation);
OutputIS.escribeSalida(ficheroSalida[1], test, entradas, salida, nEntradas, relation);
}
public void ejecutar() {
int i, j, l, m, o;
int nClases;
int claseObt;
boolean marcas[];
double conjS[][];
int clasesS[];
int eleS[], eleT[];
int bestAc, aciertos;
int temp[];
int pos, tmp;
long tiempo = System.currentTimeMillis();
/*Getting the number of different classes*/
nClases = 0;
for (i = 0; i < clasesTrain.length; i++) if (clasesTrain[i] > nClases) nClases = clasesTrain[i];
nClases++;
/*Inicialization of the flagged instance vector of the S set*/
marcas = new boolean[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) marcas[i] = false;
/*Allocate memory for the random selection*/
m = (int) ((porcentaje * datosTrain.length) / 100.0);
eleS = new int[m];
eleT = new int[datosTrain.length - m];
temp = new int[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) temp[i] = i;
/** Random distribution of elements in each set */
Randomize.setSeed(semilla);
for (i = 0; i < eleS.length; i++) {
pos = Randomize.Randint(i, datosTrain.length - 1);
tmp = temp[i];
temp[i] = temp[pos];
temp[pos] = tmp;
eleS[i] = temp[i];
}
for (i = 0; i < eleT.length; i++) {
pos = Randomize.Randint(m + i, datosTrain.length - 1);
tmp = temp[m + i];
temp[m + i] = temp[pos];
temp[pos] = tmp;
eleT[i] = temp[m + i];
}
for (i = 0; i < eleS.length; i++) marcas[eleS[i]] = true;
/*Building of the S set from the flags*/
conjS = new double[m][datosTrain[0].length];
clasesS = new int[m];
for (o = 0, l = 0; o < datosTrain.length; o++) {
if (marcas[o]) { // the instance will be evaluated
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[o][j];
}
clasesS[l] = clasesTrain[o];
l++;
}
}
/*Evaluation of the S set*/
bestAc = 0;
for (i = 0; i < datosTrain.length; i++) {
claseObt = KNN.evaluacionKNN2(k, conjS, clasesS, datosTrain[i], nClases);
if (claseObt == clasesTrain[i]) // correct clasification
bestAc++;
}
/*Body of the ENNRS algorithm. Change the S set in each iteration for instances
of the T set until get a complete sustitution*/
for (i = 0; i < n; i++) {
/*Preparation the set to interchange*/
for (j = 0; j < eleS.length; j++) {
pos = Randomize.Randint(j, eleT.length - 1);
tmp = eleT[j];
eleT[j] = eleT[pos];
eleT[pos] = tmp;
}
/*Interchange of instances*/
for (j = 0; j < eleS.length; j++) {
tmp = eleS[j];
eleS[j] = eleT[j];
eleT[j] = tmp;
marcas[eleS[j]] = true;
marcas[eleT[j]] = false;
}
/*Building of the S set from the flags*/
for (o = 0, l = 0; o < datosTrain.length; o++) {
if (marcas[o]) { // the instance will evaluate
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[o][j];
}
clasesS[l] = clasesTrain[o];
l++;
}
}
/*Evaluation of the S set*/
aciertos = 0;
for (j = 0; j < datosTrain.length; j++) {
claseObt = KNN.evaluacionKNN2(k, conjS, clasesS, datosTrain[j], nClases);
if (claseObt == clasesTrain[j]) // correct clasification
aciertos++;
}
if (aciertos > bestAc) { // keep S
bestAc = aciertos;
} else { // undo changes
for (j = 0; j < eleS.length; j++) {
tmp = eleS[j];
eleS[j] = eleT[j];
eleT[j] = tmp;
marcas[eleS[j]] = true;
marcas[eleT[j]] = false;
}
}
}
/*Building of the S set from the flags*/
/*Building of the S set from the flags*/
for (o = 0, l = 0; o < datosTrain.length; o++) {
if (marcas[o]) { // the instance will evaluate
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[o][j];
}
clasesS[l] = clasesTrain[o];
l++;
}
}
System.out.println(
"ENNRS " + relation + " " + (double) (System.currentTimeMillis() - tiempo) / 1000.0 + "s");
// COn conjS me vale.
int trainRealClass[][];
int trainPrediction[][];
trainRealClass = new int[datosTrain.length][1];
trainPrediction = new int[datosTrain.length][1];
// Working on training
for (i = 0; i < datosTrain.length; i++) {
trainRealClass[i][0] = clasesTrain[i];
trainPrediction[i][0] = KNN.evaluate(datosTrain[i], conjS, nClases, clasesS, this.k);
}
KNN.writeOutput(ficheroSalida[0], trainRealClass, trainPrediction, entradas, salida, relation);
// Working on test
int realClass[][] = new int[datosTest.length][1];
int prediction[][] = new int[datosTest.length][1];
// Check time
for (i = 0; i < realClass.length; i++) {
realClass[i][0] = clasesTest[i];
prediction[i][0] = KNN.evaluate(datosTest[i], conjS, nClases, clasesS, this.k);
}
KNN.writeOutput(ficheroSalida[1], realClass, prediction, entradas, salida, relation);
}
/**
* Evaluates a chromosome
*
* @param datos Reference to the training set
* @param real Reference to the training set (real valued)
* @param nominal Reference to the training set (nominal valued)
* @param nulos Reference to the training set (null values)
* @param clases Output attribute of each instance
* @param alfa Alpha value of the fitness function
* @param kNeigh Number of neighbors for the KNN algorithm
* @param nClases Number of classes of the problem
* @param distanceEu True= Euclidean distance; False= HVDM
*/
public void evalua(
double datos[][],
double real[][],
int nominal[][],
boolean nulos[][],
int clases[],
double alfa,
int kNeigh,
int nClases,
boolean distanceEu) {
int i, j, l, m;
int aciertos = 0;
double M, s;
double conjS[][];
double conjR[][];
int conjN[][];
boolean conjM[][];
int clasesS[];
int vecinos[];
int claseObt;
int vecinoCercano;
double dist, minDist;
M = (double) datos.length;
s = (double) genesActivos();
if (kNeigh > 1) {
vecinos = new int[kNeigh];
conjS = new double[(int) s][datos[0].length];
conjR = new double[(int) s][datos[0].length];
conjN = new int[(int) s][datos[0].length];
conjM = new boolean[(int) s][datos[0].length];
clasesS = new int[(int) s];
for (j = 0, l = 0; j < datos.length; j++) {
if (cuerpo[j]) { // the instance must be copied to the solution
for (m = 0; m < datos[j].length; m++) {
conjS[l][m] = datos[j][m];
conjR[l][m] = real[j][m];
conjN[l][m] = nominal[j][m];
conjM[l][m] = nulos[j][m];
}
clasesS[l] = clases[j];
l++;
}
}
for (i = 0; i < datos.length; i++) {
claseObt =
KNN.evaluacionKNN2(
kNeigh,
conjS,
conjR,
conjN,
conjM,
clasesS,
datos[i],
real[i],
nominal[i],
nulos[i],
nClases,
distanceEu,
vecinos);
if (claseObt >= 0) if (clases[i] == claseObt) aciertos++;
}
} else {
for (i = 0; i < datos.length; i++) {
vecinoCercano = -1;
minDist = Double.POSITIVE_INFINITY;
for (j = 0; j < datos.length; j++) {
if (cuerpo[j]) { // It is in S
dist =
KNN.distancia(
datos[i],
real[i],
nominal[i],
nulos[i],
datos[j],
real[j],
nominal[j],
nulos[j],
distanceEu);
if (dist < minDist && dist != 0) {
minDist = dist;
vecinoCercano = j;
}
}
}
if (vecinoCercano >= 0) if (clases[i] == clases[vecinoCercano]) aciertos++;
}
}
calidad = ((double) (aciertos) / M) * alfa * 100.0;
calidad += ((1.0 - alfa) * 100.0 * (M - s) / M);
cruzado = false;
} // end-method
public void ejecutar() {
int i, j, l;
boolean marcas[];
boolean marcas2[];
boolean marcastmp[];
boolean incorrect[];
int nSel;
double conjS[][];
double conjR[][];
int conjN[][];
boolean conjM[][];
int clasesS[];
Vector<Integer> vecinos[];
int next;
int maxneigh;
int pos;
int borrado;
int claseObt;
int nClases;
long tiempo = System.currentTimeMillis();
/*Getting the number of differents classes*/
nClases = 0;
for (i = 0; i < clasesTrain.length; i++) if (clasesTrain[i] > nClases) nClases = clasesTrain[i];
nClases++;
/*Inicialization of the flagged instances vector for a posterior copy*/
marcas = new boolean[datosTrain.length];
marcas2 = new boolean[datosTrain.length];
incorrect = new boolean[datosTrain.length];
marcastmp = new boolean[datosTrain.length];
Arrays.fill(marcas, true);
Arrays.fill(marcas2, true);
Arrays.fill(incorrect, false);
Arrays.fill(marcastmp, true);
vecinos = new Vector[datosTrain.length];
for (i = 0; i < datosTrain.length; i++) vecinos[i] = new Vector<Integer>();
for (i = 0; i < datosTrain.length; i++) {
next = nextNeighbour(marcas, datosTrain, i, vecinos[i]);
for (j = 0; j < datosTrain.length; j++) marcastmp[j] = marcas[j];
while (next >= 0 && clasesTrain[next] == clasesTrain[i]) {
vecinos[i].add(new Integer(next));
marcastmp[next] = false;
next = nextNeighbour(marcastmp, datosTrain, i, vecinos[i]);
}
}
maxneigh = vecinos[0].size();
pos = 0;
for (i = 1; i < datosTrain.length; i++) {
if (vecinos[i].size() > maxneigh) {
maxneigh = vecinos[i].size();
pos = i;
}
}
while (maxneigh > 0) {
for (i = 0; i < vecinos[pos].size(); i++) {
borrado = vecinos[pos].elementAt(i).intValue();
marcas[borrado] = false;
for (j = 0; j < datosTrain.length; j++) {
vecinos[j].removeElement(new Integer(borrado));
}
vecinos[borrado].clear();
}
vecinos[pos].clear();
maxneigh = vecinos[0].size();
pos = 0;
for (i = 1; i < datosTrain.length; i++) {
if (vecinos[i].size() > maxneigh) {
maxneigh = vecinos[i].size();
pos = i;
}
}
}
/*Building of the S set from the flags*/
nSel = 0;
for (i = 0; i < datosTrain.length; i++) if (marcas[i]) nSel++;
conjS = new double[nSel][datosTrain[0].length];
conjR = new double[nSel][datosTrain[0].length];
conjN = new int[nSel][datosTrain[0].length];
conjM = new boolean[nSel][datosTrain[0].length];
clasesS = new int[nSel];
for (i = 0, l = 0; i < datosTrain.length; i++) {
if (marcas[i]) { // the instance will be copied to the solution
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[i][j];
conjR[l][j] = realTrain[i][j];
conjN[l][j] = nominalTrain[i][j];
conjM[l][j] = nulosTrain[i][j];
}
clasesS[l] = clasesTrain[i];
l++;
}
}
for (i = 0; i < datosTrain.length; i++) {
/*Apply 1-NN to the instance*/
claseObt =
KNN.evaluacionKNN2(
1,
conjS,
conjR,
conjN,
conjM,
clasesTrain,
datosTrain[i],
realTrain[i],
nominalTrain[i],
nulosTrain[i],
nClases,
true);
if (claseObt != clasesTrain[i]) {
incorrect[i] = true;
}
}
for (i = 0; i < datosTrain.length; i++) vecinos[i] = new Vector<Integer>();
for (i = 0; i < datosTrain.length; i++) {
if (incorrect[i]) {
next = nextNeighbour(marcas2, datosTrain, i, vecinos[i]);
for (j = 0; j < datosTrain.length; j++) marcastmp[j] = marcas2[j];
while (next >= 0 && clasesTrain[next] == clasesTrain[i]) {
vecinos[i].add(new Integer(next));
marcastmp[next] = false;
next = nextNeighbour(marcastmp, datosTrain, i, vecinos[i]);
}
}
}
maxneigh = vecinos[0].size();
pos = 0;
for (i = 1; i < datosTrain.length; i++) {
if (vecinos[i].size() > maxneigh) {
maxneigh = vecinos[i].size();
pos = i;
}
}
while (maxneigh > 0) {
for (i = 0; i < vecinos[pos].size(); i++) {
borrado = vecinos[pos].elementAt(i).intValue();
marcas2[borrado] = false;
for (j = 0; j < datosTrain.length; j++) {
vecinos[j].removeElement(new Integer(borrado));
}
vecinos[borrado].clear();
}
vecinos[pos].clear();
maxneigh = vecinos[0].size();
pos = 0;
for (i = 1; i < datosTrain.length; i++) {
if (vecinos[i].size() > maxneigh) {
maxneigh = vecinos[i].size();
pos = i;
}
}
}
for (i = 0; i < marcas.length; i++) marcas[i] |= (marcas2[i] & incorrect[i]);
/*Building of the S set from the flags*/
nSel = 0;
for (i = 0; i < datosTrain.length; i++) if (marcas[i]) nSel++;
conjS = new double[nSel][datosTrain[0].length];
conjR = new double[nSel][datosTrain[0].length];
conjN = new int[nSel][datosTrain[0].length];
conjM = new boolean[nSel][datosTrain[0].length];
clasesS = new int[nSel];
for (i = 0, l = 0; i < datosTrain.length; i++) {
if (marcas[i]) { // the instance will be copied to the solution
for (j = 0; j < datosTrain[0].length; j++) {
conjS[l][j] = datosTrain[i][j];
conjR[l][j] = realTrain[i][j];
conjN[l][j] = nominalTrain[i][j];
conjM[l][j] = nulosTrain[i][j];
}
clasesS[l] = clasesTrain[i];
l++;
}
}
System.out.println(
"Reconsistent "
+ relation
+ " "
+ (double) (System.currentTimeMillis() - tiempo) / 1000.0
+ "s");
// COn conjS me vale.
int trainRealClass[][];
int trainPrediction[][];
trainRealClass = new int[datosTrain.length][1];
trainPrediction = new int[datosTrain.length][1];
// Working on training
for (i = 0; i < datosTrain.length; i++) {
trainRealClass[i][0] = clasesTrain[i];
trainPrediction[i][0] = KNN.evaluate(datosTrain[i], conjS, nClases, clasesS, 1);
}
KNN.writeOutput(ficheroSalida[0], trainRealClass, trainPrediction, entradas, salida, relation);
// Working on test
int realClass[][] = new int[datosTest.length][1];
int prediction[][] = new int[datosTest.length][1];
// Check time
for (i = 0; i < realClass.length; i++) {
realClass[i][0] = clasesTest[i];
prediction[i][0] = KNN.evaluate(datosTest[i], conjS, nClases, clasesS, 1);
}
KNN.writeOutput(ficheroSalida[1], realClass, prediction, entradas, salida, relation);
}
public void ejecutar() {
double conjS[][];
double conjR[][];
int conjN[][];
boolean conjM[][];
int clasesS[];
int S[]; /* Binary Vector, to decide if the instance will be included*/
int i, j, l, cont;
int nClases;
int tamS;
int transformations;
int claseObt[];
int clasePredominante;
long tiempo = System.currentTimeMillis();
transformations = 0;
/*Getting the number of different classes*/
nClases = 0;
for (i = 0; i < clasesTrain.length; i++) if (clasesTrain[i] > nClases) nClases = clasesTrain[i];
nClases++;
if (nClases < 2) {
System.err.println("Input dataset is empty");
nClases = 0;
}
/*Algorithm body.
First, S=TS.
Then, for each instance of TS, the first step is to repeat the aplication of the k-nn, and then
we decide if we need to change the label of the instance or we don't need it.
*/
/*Inicialization of the candidates set, S=X, where X is the original Training Set*/
S = new int[datosTrain.length];
for (i = 0; i < S.length; i++) S[i] = 1; /* All included*/
tamS = datosTrain.length;
System.out.print("K= " + k + "\n");
System.out.print("K'= " + k2 + "\n");
for (i = 0; i < datosTrain.length; i++) {
/* I need find the k-nn of i in X - {i}, so I make conjS without i*/
conjS = new double[datosTrain.length - 1][datosTrain[0].length];
conjR = new double[datosTrain.length - 1][datosTrain[0].length];
conjN = new int[datosTrain.length - 1][datosTrain[0].length];
conjM = new boolean[datosTrain.length - 1][datosTrain[0].length];
clasesS = new int[datosTrain.length - 1];
cont = 0;
for (j = 0; j < datosTrain.length; j++) {
if (i != j) {
for (l = 0; l < datosTrain[0].length; l++) {
conjS[cont][l] = datosTrain[j][l];
conjR[cont][l] = realTrain[j][l];
conjN[cont][l] = nominalTrain[j][l];
conjM[cont][l] = nulosTrain[j][l];
}
clasesS[cont] = clasesTrain[j];
cont++;
}
}
/*Do KNN to the instance*/
claseObt =
KNN.evaluacionKNN3(
k,
conjS,
conjR,
conjN,
conjM,
clasesS,
datosTrain[i],
realTrain[i],
nominalTrain[i],
nulosTrain[i],
nClases,
distanceEu);
/*
System.out.print("Las clases de los k vecinos m�s cercanos son\n");
for(int m=0;m<k;m++){
System.out.print(claseObt[m]+ " ");
}
System.out.print("\n-----------------------------------------------\n");
*/
/*Now, we must check that we have at least k2 neighboors with the same class. */
int max = 0;
clasePredominante = 0;
for (int m = 0; m < claseObt.length; m++) {
int claseDeInstancia = claseObt[m]; // Select one class.
int iguales = 0;
for (j = 0; j < claseObt.length; j++) { // Check numbers of instances with this class
if (j != m) { // I can't count the same.
if (claseObt[j] == claseDeInstancia) {
iguales++;
}
}
}
// I must check if there is another class with more instances.
if (iguales > max) {
max = iguales;
clasePredominante = claseObt[m];
}
}
// System.out.print("max " + max +"\n");
// System.out.print("Clase Predominante: "+clasePredominante+"\n");
/* Max+1 = number of neighbours with the same class*/
if ((max) >= k2) {
/* if there are at least k2 neighbour, we change the class in S, */
if (clasePredominante != clasesTrain[i]) transformations++;
clasesTrain[i] = clasePredominante;
S[i] = 1;
} else {
/* Discard.*/
tamS--;
S[i] = 0;
}
}
System.out.print("S size resultante= " + tamS + "\n");
System.out.print("Transformations = " + transformations + "\n");
/*Construction of the S set from the previous vector S*/
conjS = new double[tamS][datosTrain[0].length];
conjR = new double[tamS][datosTrain[0].length];
conjN = new int[tamS][datosTrain[0].length];
conjM = new boolean[tamS][datosTrain[0].length];
clasesS = new int[tamS];
cont = 0; /* To establish the sets' sizes */
for (j = 0; j < datosTrain.length; j++) {
if (S[j] == 1) {
/* Checking the instance is included*/
for (l = 0; l < datosTrain[0].length; l++) {
conjS[cont][l] = datosTrain[j][l];
conjR[cont][l] = realTrain[j][l];
conjN[cont][l] = nominalTrain[j][l];
conjM[cont][l] = nulosTrain[j][l];
}
clasesS[cont] = clasesTrain[j];
cont++;
}
}
System.out.println(
"Time elapse: " + (double) (System.currentTimeMillis() - tiempo) / 1000.0 + "s");
OutputIS.escribeSalida(
ficheroSalida[0], conjR, conjN, conjM, clasesS, entradas, salida, nEntradas, relation);
OutputIS.escribeSalida(ficheroSalida[1], test, entradas, salida, nEntradas, relation);
}