修改周期行计算方式

src/main/java/cn/mesalab/service/BaselineSingleThread.java

@@ -39,7 +39,7 @@ public class BaselineSingleThread extends Thread {
     private final CountDownLatch countDownLatch;
 
     private final ArrayList<Integer> frequencyBinCounter = new ArrayList<>(Collections.nCopies(ApplicationConfig.MONITOR_FREQUENCY_BIN_NUM, 0));
-    private final ArrayList<Integer> generateTypeCounter = new ArrayList<>(Collections.nCopies(3, 0));
+    private final ArrayList<Integer> generateTypeCounter = new ArrayList<>(Collections.nCopies(2, 0));
     private int discardBaselineCounter = 0;
     private ArrayList<String> discardIpList = new ArrayList<>();
 

src/main/java/cn/mesalab/utils/SeriesUtils.java

@@ -3,10 +3,13 @@ package cn.mesalab.utils;
 import cn.mesalab.config.ApplicationConfig;
 import cn.mesalab.dao.DruidData;
 import com.google.common.collect.Lists;
+import com.sun.istack.internal.NotNull;
 import org.jfree.util.Log;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
+import org.springframework.lang.NonNull;
 
+import javax.annotation.Nonnull;
 import java.io.BufferedReader;
 import java.io.FileReader;
 import java.time.Duration;
@@ -21,9 +24,6 @@ import java.util.*;
  */
 public class SeriesUtils {
     private static final Logger LOG = LoggerFactory.getLogger(SeriesUtils.class);
-
-    private static DruidData druidData = new DruidData();
-
     public static List<Map<String, Object>> readCsvToList(String filePath) {
         List<Map<String, Object>> list = new ArrayList<Map<String, Object>>();
 
@@ -100,38 +100,39 @@ public class SeriesUtils {
      * @return
      */
     public static Boolean isPeriod(List<Integer> historicalSeries){
-        Boolean result = true;
+        boolean result = true;
         List<List<Integer>> partitions = Lists.partition(historicalSeries, 24*60/ApplicationConfig.HISTORICAL_GRAD);
-        List<Integer> aggregatedPart = Arrays.asList();
-        try{
-            aggregatedPart = columnAverage(partitions.subList(0, ApplicationConfig.READ_HISTORICAL_DAYS-1));
-        } catch (IndexOutOfBoundsException e){
-            Log.error("历史");
-        }
 
         // Pearson corrcoef
-        double pearsonCorrelationScore = getPearsonCorrelationScore(aggregatedPart.stream().mapToInt(Integer::valueOf).toArray(),
-                partitions.get(partitions.size() - 1).stream().mapToInt(Integer::valueOf).toArray());
-
+        int size = partitions.size();
+        ArrayList<Double> corrScores = new ArrayList<>();
+        for(int i = 0; i < size; i ++){
+            for(int j = i+1; j < size; j ++){
+                corrScores.add(getPearsonCorrelationScore(partitions.get(i).stream().mapToDouble(Number::doubleValue).toArray(),
+                        partitions.get(j).stream().mapToDouble(Integer::valueOf).toArray()));
+            }
+        }
+        double pearsonCorrelationScore = corrScores.stream().mapToDouble(Number::doubleValue).average().orElse(0.0);
         if (pearsonCorrelationScore < ApplicationConfig.BASELINE_PERIOD_CORR_THRE){
             result=false;
         }
         return result;
     }
 
-    public static double getPearsonCorrelationScore(int[] xData, int[] yData) {
+    public static double getPearsonCorrelationScore(double[] xData, double[] yData) {
+
         if (xData.length != yData.length) {
             Log.error("Pearson CorrelationScore 数组长度不相等!");
         }
-        int xMeans;
-        int yMeans;
+        double xMeans;
+        double yMeans;
         double numerator = 0;
         double denominator = 0;
 
         double result = 0;
         // 拿到两个数据的平均值
-        xMeans = (int) getMeans(xData);
-        yMeans = (int) getMeans(yData);
+        xMeans = getMeans(xData);
+        yMeans = getMeans(yData);
         // 计算皮尔逊系数的分子
         numerator = generateNumerator(xData, xMeans, yData, yMeans);
         // 计算皮尔逊系数的分母
@@ -143,7 +144,7 @@ public class SeriesUtils {
         return result;
     }
 
-    private static int generateNumerator(int[] xData, int xMeans, int[] yData, int yMeans) {
+    private static float generateNumerator(double[] xData, double xMeans, double[] yData, double yMeans) {
         int numerator = 0;
         for (int i = 0; i < xData.length; i++) {
             numerator += (xData[i] - xMeans) * (yData[i] - yMeans);
@@ -151,22 +152,22 @@ public class SeriesUtils {
         return numerator;
     }
 
-    private static double generateDenomiator(int[] xData, int xMeans, int[] yData, int yMeans) {
+    private static double generateDenomiator(double[] xData, double xMeans, double[] yData, double yMeans) {
         double xSum = 0.0;
-        for (int i = 0; i < xData.length; i++) {
-            xSum += (xData[i] - xMeans) * (xData[i] - xMeans);
+        for (double xDatum : xData) {
+            xSum += (xDatum - xMeans) * (xDatum - xMeans);
         }
         double ySum = 0.0;
-        for (int i = 0; i < yData.length; i++) {
-            ySum += (yData[i] - yMeans) * (yData[i] - yMeans);
+        for (double yDatum : yData) {
+            ySum += (yDatum - yMeans) * (yDatum - yMeans);
         }
         return Math.sqrt(xSum) * Math.sqrt(ySum);
     }
 
-    private static double getMeans(int[] datas) {
-        double sum = 0.0;
-        for (int i = 0; i < datas.length; i++) {
-            sum += datas[i];
+    private static double getMeans(double[] datas) {
+        float sum = 0;
+        for (double data : datas) {
+            sum += data;
         }
         return sum / datas.length;
     }
@@ -175,8 +176,8 @@ public class SeriesUtils {
         ArrayList<Integer> averages = new ArrayList<>();
         for(int i=0; i<list.get(0).size(); i++){
             int columnSum = 0;
-            for(int j = 0; j< list.size(); j++){
-                columnSum += list.get(j).get(i);
+            for (List<Integer> integers : list) {
+                columnSum += integers.get(i);
             }
             averages.add(columnSum / list.size());
         }