我试图将一组数据的进度和出度求和。 这是示例数据:
Source Target
1 2
2 1
3 1
2 3
所以预期的输出是:
ID In degree Out degree
1 2 1
2 1 2
3 1 1
如何使用mapreduce Java做到这一点,并在一行中打印出结果。
答案 0 :(得分:0)
涉及一项MR工作的一种选择: 假设原始数据集看起来像[node1,node2]:
-mapper读取原始数据集,并为每行发出三元组[node1,node2,out]和[node2,node1,in]
-reducer以[key,node,label]的形式从映射器获取三元组,通过分别计算每个密钥的“ out”标签和“ in”标签来计算outdegree和indegree,并以[key,indegree,学位]
实现与下面类似(假设数据集中的node1
和node2
用空格隔开,并且还假设数据集仅包含不同的对):
映射器:
public class YourMapper extends MapReduceBase implements Mapper<LongWritable, Text, Text, Text> {
public void map(LongWritable key, Text value, OutputCollector<Text, Text> output, Reporter reporter) throws IOException {
String line = value.toString();
String[] line_spl = line.split(" ");
String node1 = line_spl[0];
String node2 = line_spl[1];
Text node1_txt = new Text(node1);
Text node2_txt = new Text(node2);
Text emit_out = new Text("out_" + node2);
Text emit_in = new Text("in_" + node1);
output.collect(node1_txt, emit_out);
output.collect(node2_txt, emit_in );
}//end map function
}//end mapper class
减速器:
public class YourReducer extends MapReduceBase implements Reducer<Text, Text, Text, Text> {
public void reduce(Text key, Iterator<Text> values, OutputCollector<Text, Text> output, Reporter reporter) throws IOException {
int count_outs = 0;
int count_ins = 0;
while (values.hasNext()) {
Text value = (Text) values.next();
String value_str = value.toString();
if(value_str.startsWith("out_"))
count_outs++;
else
if(value_str.startsWith("in_"))
count_ins++;
}
Text out = new Text(count_ins + " " + count_outs);
output.collect(key, out);
}//end reduce function
}//end reducer class