java语言实现权重随机算法 java语言实现权重随机算法完整实例
buqutianya 人气:0想了解java语言实现权重随机算法完整实例的相关内容吗,buqutianya在本文为您仔细讲解java语言实现权重随机算法的相关知识和一些Code实例,欢迎阅读和指正,我们先划重点:java,权重随机算法,java,权重抽奖随机算法,java,按权重分配算法,java,权重算法,下面大家一起来学习吧。
前言
现在app就是雨后春笋,嗖嗖的往外冒啊,有经验的、没经验的、有资历的、没资历的都想着创业,创业的90%以上都要做一个app出来,好像成了创业的标配。
做了app就得推广啊,怎么推,发券送钱是最多用的被不可少的了,现在好多产品或者运营都要求能够随机出优惠券的金额,但是呢又不能过于随机,送出去的券都是钱吗,投资人的钱,是吧。
所以,在随机生成的金额中就要求,小额度的几率要大,大额度的几率要小,比如说3元的70%,5块的25%,10块的5%,这个样子的概率去生成优惠券,这个怎么办呢?
对于上述的问题,直接用我们的Random.next(Integer range);就不够了。因为这个伪随机不带权重,3,5,10出现的概率都是一样的。
实现思路
还是拿上述的例子,3出现的概率是70%,我们给他的权重赋值为70,5出现的概率为25%,我们给他的权重赋值为25,10出现的概率为5%,我们给他的权重赋值为5.
我们按照顺序计算出权重的加和,把当前数字出现的权重加和前的值作为其权重范围的起点值,把加和后的值作为其权重范围的终点值。
这样的话,我们就可以使用Random.next(100)来做随机数,然后判断随机数落在的范围,然后映射到对应的优惠券数值即可。
java实现
package com.nggirl.test.weight.random; import java.util.ArrayList; import java.util.HashMap; import java.util.List; import java.util.Random; public class WeightRandom { public static void main(String[] args){ WeightRandom wr = new WeightRandom(); wr.initWeight(new String[]{ "1","2","3","4" } , new Integer[]{ 100,100,200,600 } ); Random r = new Random(); for (int i = 0; i < 10; i++){ Integer rv = r.nextint(wr.getMaxRandomValue()); System.out.println(rv); System.out.println(wr.getElementByRandomValue(rv).getKey() + " " + rv); } HashMap<String, Integer> keyCount = new HashMap<String, Integer>(); keyCount.put("1", 0); keyCount.put("2", 0); keyCount.put("3", 0); keyCount.put("4", 0); for (int i = 0; i < 10000; i++){ Integer rv = r.nextint(wr.getMaxRandomValue()); String key = wr.getElementByRandomValue(rv).getKey(); keyCount.put(key, keyCount.get(key).intValue()+1); } System.out.println(""); } private List<WeightElement> weightElements; public void initWeight(String[] keys, Integer[] weights){ if(keys == null || weights == null || keys.length != weights.length){ return; } weightElements = new ArrayList<WeightElement>(); for (int i=0; i< keys.length; i++){ weightElements.add(new WeightElement(keys[i], weights[i])); } rangeWeightElemnts(); printRvs(); } private void rangeWeightElemnts(){ if(weightElements.size() == 0){ return; } WeightElement ele0 = weightElements.get(0); ele0.setThresholdLow(0); ele0.setThresholdHigh(ele0.getWeight()); for (int i = 1; i < weightElements.size(); i++){ WeightElement curElement = weightElements.get(i); WeightElement preElement = weightElements.get(i - 1); curElement.setThresholdLow(preElement.getThresholdHigh()); curElement.setThresholdHigh(curElement.getThresholdLow() + curElement.getWeight()); } } public WeightElement getElementByRandomValue(Integer rv){ //因为元素权重范围有序递增,所以这里可以改为二分查找 for (WeightElement e:weightElements){ if(rv >= e.getThresholdLow() && rv < e.getThresholdHigh()){ return e; } } return null; } public Integer getMaxRandomValue(){ if(weightElements == null || weightElements.size() == 0){ return null; } return weightElements.get(weightElements.size() - 1).getThresholdHigh(); } public void printRvs(){ for (WeightElement e:weightElements){ System.out.println(e.toString()); } } static class WeightElement{ /** * 元素标记 */ private String key; /** * 元素权重 */ private Integer weight; /** * 权重对应随机数范围低线 */ private Integer thresholdLow; /** * 权重对应随机数范围高线 */ private Integer thresholdHigh; public WeightElement(){ } public WeightElement(Integer weight){ this.key = weight.toString(); this.weight = weight; } public WeightElement(String key, Integer weight){ this.key = key; this.weight = weight; } public String getKey() { return key; } public void setKey(String key) { this.key = key; } public Integer getWeight() { return weight; } public void setWeight(Integer weight) { this.weight = weight; } public Integer getThresholdLow() { return thresholdLow; } public void setThresholdLow(Integer thresholdLow) { this.thresholdLow = thresholdLow; } public Integer getThresholdHigh() { return thresholdHigh; } public void setThresholdHigh(Integer thresholdHigh) { this.thresholdHigh = thresholdHigh; } public String toString(){ return "key:"+this.key + " weight:" + this.weight + " low:"+this.thresholdLow+" heigh:"+this.thresholdHigh; } } }
结果:
2 102 876 4 876
二分法的实现
public WeightElement getElementByRandomValue(Integer rv){ if(rv < 0 || rv > getMaxRandomValue()-1){ return null; } //此时rv必然在0 - getMaxRandomValue()-1范围内, //也就是必然能够命中某一个值 int start = 0, end = weightElements.size() - 1; int index = weightElements.size()/2; while(true){ if(rv < weightElements.get(index).getThresholdLow()){ end = index - 1; } else if(rv >= weightElements.get(index).getThresholdHigh()){ start = index + 1; } else{ return weightElements.get(index); } index = (start + end)/2; } }
下面再分享一则实例,加强对权重随机算法的理解,一次到位!
权重随机算法在抽奖,资源调度等系统中应用还是比较广泛的,一个简单的按照权重来随机的实现,权重为几个随机对象(分类)的命中的比例,权重设置越高命中越容易,之和可以不等于100;
简单实现代码如下:
import java.util.ArrayList; import java.util.List; import java.util.Random; public class WeightRandom { static List<WeightCategory> categorys = new ArrayList<WeightCategory>(); private static Random random = new Random(); public static void initData() { WeightCategory wc1 = new WeightCategory("A",60); WeightCategory wc2 = new WeightCategory("B",20); WeightCategory wc3 = new WeightCategory("C",20); categorys.add(wc1); categorys.add(wc2); categorys.add(wc3); } public static void main(String[] args) { initData(); Integer weightSum = 0; for (WeightCategory wc : categorys) { weightSum += wc.getWeight(); } if (weightSum <= 0) { System.err.println("Error: weightSum=" + weightSum.toString()); return; } Integer n = random.nextint(weightSum); // n in [0, weightSum) Integer m = 0; for (WeightCategory wc : categorys) { if (m <= n && n < m + wc.getWeight()) { System.out.println("This Random Category is "+wc.getCategory()); break; } m += wc.getWeight(); } } } class WeightCategory { private String category; private Integer weight; public WeightCategory() { super(); } public WeightCategory(String category, Integer weight) { super(); this.setCategory(category); this.setWeight(weight); } public Integer getWeight() { return weight; } public void setWeight(Integer weight) { this.weight = weight; } public String getCategory() { return category; } public void setCategory(String category) { this.category = category; } }
结果:
总结
以上就是本文关于java语言实现权重随机算法完整实例的全部内容,希望对大家有所帮助。如有不足之处,欢迎留言指出。感谢朋友们对本站的支持!
加载全部内容