MapReduce業務 - 圖片關聯計算

哥不是小蘿莉發表於2015-09-02

1.概述

  最近在和人交流時談到資料相似度和資料共性問題,而剛好在業務層面有類似的需求,今天和大家分享這類問題的解決思路,分享目錄如下所示:

  • 業務背景
  • 編碼實踐
  • 預覽截圖

  下面開始今天的內容分享。

2.業務背景

  目前有這樣一個背景,在一大堆資料中,裡面存放著圖片的相關資訊,如下圖所示:

  上圖只是給大家列舉的一個示例資料格式,第一列表示自身圖片,第二、第三......等列表示與第一列相關聯的圖片資訊。那麼我們從這堆資料中如何找出他們擁有相同圖片資訊的圖片。

2.1 實現思路

  那麼,我們在明確了上述需求後,下面我們來分析它的實現思路。首先,我們通過上圖所要實現的目標結果,其最終計算結果如下所示:

pic_001pic_002 pic_003,pic_004,pic_005
pic_001pic_003 pic_002,pic_005
pic_001pic_004 pic_002,pic_005
pic_001pic_005 pic_002,pic_003,pic_004
......

  結果如上所示,找出兩兩圖片之間的共性圖片,結果未列完整,只是列舉了部分,具體結果大家可以參考截圖預覽的相關資訊。

  下面給大家介紹解決思路,通過觀察資料,我們可以發現在上述資料當中,我們要計算圖片兩兩的共性圖片,可以從關聯圖片入手,在關聯圖片中我們可以找到共性圖片的關聯資訊,比如:我們要計算pic001pic002圖片的共性圖片,我們可以在關聯圖片中找到兩者(pic001pic002組合)後對應的自身圖片(key),最後在將所有的key求並集即為兩者的共性圖片資訊,具體資訊如下圖所示:

  通過上圖,我們可以知道具體的實現思路,步驟如下所示:

  • 第一步:拆分資料,關聯資料兩兩組合作為Key輸出。
  • 第二步:將相同Key分組,然後求並集得到計算結果。

  這裡使用一個MR來完成此項工作,在明白了實現思路後,我們接下來去實現對應的編碼。

3.編碼實踐

  • 拆分資料,兩兩組合。
public static class PictureMap extends Mapper<LongWritable, Text, Text, Text> {

        @Override
        protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, Text>.Context context)
                throws IOException, InterruptedException {
            StringTokenizer strToken = new StringTokenizer(value.toString());
            Text owner = new Text();

            Set<String> set = new TreeSet<String>();

            owner.set(strToken.nextToken());
            while (strToken.hasMoreTokens()) {
                set.add(strToken.nextToken());
            }

            String[] relations = new String[set.size()];
            relations = set.toArray(relations);

            for (int i = 0; i < relations.length; i++) {
                for (int j = i + 1; j < relations.length; j++) {
                    String outPutKey = relations[i] + relations[j];
                    context.write(new Text(outPutKey), owner);
                }

            }
        }
    }
  • 按Key分組,求並集
public static class PictureReduce extends Reducer<Text, Text, Text, Text> {

        @Override
        protected void reduce(Text key, Iterable<Text> values, Reducer<Text, Text, Text, Text>.Context context)
                throws IOException, InterruptedException {
            String common = "";
            for (Text val : values) {
                if (common == "") {
                    common = val.toString();
                } else {
                    common = common + "," + val.toString();
                }
            }
            context.write(key, new Text(common));
        }
    }
  • 完整示例
package cn.hadoop.hdfs.example;

import java.io.IOException;
import java.util.Set;
import java.util.StringTokenizer;
import java.util.TreeSet;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.conf.Configured;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.util.Tool;
import org.apache.hadoop.util.ToolRunner;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import cn.hadoop.hdfs.util.HDFSUtils;
import cn.hadoop.hdfs.util.SystemConfig;

/**
 * @Date Aug 31, 2015
 *
 * @Author dengjie
 *
 * @Note Find picture relations
 */
public class PictureRelations extends Configured implements Tool {

    private static Logger log = LoggerFactory.getLogger(PictureRelations.class);
    private static Configuration conf;

    static {
        String tag = SystemConfig.getProperty("dev.tag");
        String[] hosts = SystemConfig.getPropertyArray(tag + ".hdfs.host", ",");
        conf = new Configuration();
        conf.set("fs.defaultFS", "hdfs://cluster1");
        conf.set("dfs.nameservices", "cluster1");
        conf.set("dfs.ha.namenodes.cluster1", "nna,nns");
        conf.set("dfs.namenode.rpc-address.cluster1.nna", hosts[0]);
        conf.set("dfs.namenode.rpc-address.cluster1.nns", hosts[1]);
        conf.set("dfs.client.failover.proxy.provider.cluster1",
                "org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider");
        conf.set("fs.hdfs.impl", org.apache.hadoop.hdfs.DistributedFileSystem.class.getName());
        conf.set("fs.file.impl", org.apache.hadoop.fs.LocalFileSystem.class.getName());
    }

    public static class PictureMap extends Mapper<LongWritable, Text, Text, Text> {

        @Override
        protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, Text>.Context context)
                throws IOException, InterruptedException {
            StringTokenizer strToken = new StringTokenizer(value.toString());
            Text owner = new Text();

            Set<String> set = new TreeSet<String>();

            owner.set(strToken.nextToken());
            while (strToken.hasMoreTokens()) {
                set.add(strToken.nextToken());
            }

            String[] relations = new String[set.size()];
            relations = set.toArray(relations);

            for (int i = 0; i < relations.length; i++) {
                for (int j = i + 1; j < relations.length; j++) {
                    String outPutKey = relations[i] + relations[j];
                    context.write(new Text(outPutKey), owner);
                }

            }
        }
    }

    public static class PictureReduce extends Reducer<Text, Text, Text, Text> {

        @Override
        protected void reduce(Text key, Iterable<Text> values, Reducer<Text, Text, Text, Text>.Context context)
                throws IOException, InterruptedException {
            String common = "";
            for (Text val : values) {
                if (common == "") {
                    common = val.toString();
                } else {
                    common = common + "," + val.toString();
                }
            }
            context.write(key, new Text(common));
        }
    }

    public int run(String[] args) throws Exception {
        final Job job = Job.getInstance(conf);
        job.setJarByClass(PictureMap.class);
        job.setMapperClass(PictureMap.class);
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Text.class);
        job.setReducerClass(PictureReduce.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(Text.class);
        FileInputFormat.setInputPaths(job, args[0]);
        FileOutputFormat.setOutputPath(job, new Path(args[1]));
        int status = job.waitForCompletion(true) ? 0 : 1;
        return status;
    }

    public static void main(String[] args) {
        try {
            if (args.length != 1) {
                log.warn("args length must be 1 and as date param");
                return;
            }
            String tmpIn = SystemConfig.getProperty("hdfs.input.path.v2");
            String tmpOut = SystemConfig.getProperty("hdfs.output.path.v2");
            String inPath = String.format(tmpIn, "t_pic_20150801.log");
            String outPath = String.format(tmpOut, "meta/" + args[0]);

            // bak dfs file to old
            HDFSUtils.bak(tmpOut, outPath, "meta/" + args[0] + "-old", conf);

            args = new String[] { inPath, outPath };
            int res = ToolRunner.run(new Configuration(), new PictureRelations(), args);
            System.exit(res);
        } catch (Exception ex) {
            ex.printStackTrace();
            log.error("Picture relations task has error,msg is" + ex.getMessage());
        }

    }

}

4.截圖預覽

  關於計算結果,如下圖所示:

5.總結

  本篇部落格只是從思路上實現了圖片關聯計算,在資料量大的情況下,是有待優化的,這裡就不多做贅述了,後續有時間在為大家分析其中的細節。

6.結束語

  這篇部落格就和大家分享到這裡,如果大家在研究學習的過程當中有什麼問題,可以加群進行討論或傳送郵件給我,我會盡我所能為您解答,與君共勉!

相關文章