瀏覽代碼

dssm train

丁云鹏 4 月之前
父節點
當前提交
2440980089

+ 2 - 1
recommend-model-produce/src/main/java/com/tzld/piaoquan/recommend/model/produce/i2i/I2IDSSMService.java

@@ -39,6 +39,7 @@ public class I2IDSSMService {
         CMDService cmd = new CMDService();
         Map<String, String> argMap = cmd.parse(args);
         String file = argMap.get("path");
+        int repartition = NumberUtils.toInt(argMap.get("repartition"), 64);
 
         // 加载模型
         String bucketName = "art-recommend";
@@ -73,7 +74,7 @@ public class I2IDSSMService {
 
         // 将处理后的数据写入新的文件,使用Gzip压缩
         String outputPath = "hdfs:/dyp/vec2";
-        processedRdd.saveAsTextFile(outputPath, GzipCodec.class);
+        processedRdd.coalesce(repartition).saveAsTextFile(outputPath, GzipCodec.class);
     }
 
     private String processLine(String line, Predictor predictor) {