Add File

src/main/java/org/dromara/easyai/tools/LVQ.java

@@ -0,0 +1,146 @@
+package org.dromara.easyai.tools;
+
+import org.dromara.easyai.matrixTools.Matrix;
+import org.dromara.easyai.matrixTools.MatrixOperation;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Random;
+
+/**
+ * @author lidapeng
+ * @description LVQ 向量量化
+ * @date 5:36 下午 2020/2/4
+ */
+public class LVQ {
+    private int typeNub;//原型聚类个数,即分类个数(需要模型返回)
+    private MatrixBody[] model;//原型向量(需要模型返回)
+    private final List<MatrixBody> matrixList = new ArrayList<>();
+    private final float studyPoint;//量化学习率
+    private int length;//向量长度(需要返回)
+    private boolean isReady = false;
+    private final int lvqNub;
+    private final MatrixOperation matrixOperation = new MatrixOperation();
+
+    public void setTypeNub(int typeNub) {
+        this.typeNub = typeNub;
+    }
+
+    public void setModel(MatrixBody[] model) {
+        this.model = model;
+    }
+
+    public void setLength(int length) {
+        this.length = length;
+    }
+
+    public void setReady(boolean ready) {
+        isReady = ready;
+    }
+
+    public boolean isReady() {
+        return isReady;
+    }
+
+    public int getTypeNub() {
+        return typeNub;
+    }
+
+    public int getLength() {
+        return length;
+    }
+
+    public LVQ(int typeNub, int lvqNub, float studyPoint) {
+        this.typeNub = typeNub;
+        this.lvqNub = lvqNub;
+        this.studyPoint = studyPoint;
+        model = new MatrixBody[typeNub];
+    }
+
+    public MatrixBody[] getModel() throws Exception {
+        if (!isReady) {
+            throw new Exception("not study");
+        }
+        return model;
+    }
+
+    public void insertMatrixBody(MatrixBody matrixBody) throws Exception {
+        if (matrixBody.getMatrix().isVector() && matrixBody.getMatrix().isRowVector()) {
+            Matrix matrix = matrixBody.getMatrix();
+            if (matrixList.size() == 0) {
+                matrixList.add(matrixBody);
+                length = matrix.getY();
+            } else {
+                if (length == matrix.getY()) {
+                    matrixList.add(matrixBody);
+                } else {
+                    throw new Exception("vector length is different");
+                }
+            }
+        } else {
+            throw new Exception("this matrix is not vector or rowVector");
+        }
+    }
+
+    private void study() throws Exception {
+        for (MatrixBody matrixBody : matrixList) {
+            Matrix matrix = matrixBody.getMatrix();//特征向量
+            long type = matrixBody.getId();//类别
+            float distEnd = 0;
+            int id = 0;
+            for (int i = 0; i < typeNub; i++) {
+                MatrixBody modelBody = model[i];
+                Matrix modelMatrix = modelBody.getMatrix();
+                //修正矩阵与原矩阵的范数差
+                float dist = vectorEqual(modelMatrix, matrix);
+                if (distEnd == 0 || dist < distEnd) {
+                    id = modelBody.getId();
+                    distEnd = dist;
+                }
+            }
+            MatrixBody modelBody = model[id];
+            Matrix modelMatrix = modelBody.getMatrix();
+            boolean isRight = id == type;
+            Matrix matrix1 = op(matrix, modelMatrix, isRight);
+            modelBody.setMatrix(matrix1);
+        }
+    }
+
+    //比较两个向量之间的范数差
+    public float vectorEqual(Matrix matrix1, Matrix matrix2) throws Exception {
+        Matrix matrix = matrixOperation.sub(matrix1, matrix2);
+        return matrixOperation.getNorm(matrix);
+    }
+
+    private Matrix op(Matrix matrix, Matrix modelMatrix, boolean isRight) throws Exception {
+        Matrix matrix1 = matrixOperation.sub(matrix, modelMatrix);
+        matrixOperation.mathMul(matrix1, studyPoint);
+        Matrix matrix2;
+        if (isRight) {
+            matrix2 = matrixOperation.add(modelMatrix, matrix1);
+        } else {
+            matrix2 = matrixOperation.sub(modelMatrix, matrix1);
+        }
+        return matrix2;
+    }
+
+    public void start() throws Exception {//开始向量量化聚类
+        Random random = new Random();
+        for (int i = 0; i < typeNub; i++) {
+            MatrixBody matrixBody = new MatrixBody();
+            Matrix matrix = new Matrix(1, length);
+            matrixBody.setMatrix(matrix);
+            matrixBody.setId(i);
+            for (int j = 0; j < length; j++) {
+                matrix.setNub(0, j, random.nextFloat());
+            }
+            model[i] = matrixBody;
+        }
+        //初始化完成
+        for (int i = 0; i < lvqNub; i++) {
+            //System.out.println("================================");
+            study();
+        }
+        isReady = true;
+    }
+}