zhangbo 1 سال پیش
والد
کامیت
2932b81263
1فایلهای تغییر یافته به همراه4 افزوده شده و 0 حذف شده
  1. 4 0
      src/main/scala/com/aliyun/odps/spark/examples/makedata/makedata_02_writeredis.scala

+ 4 - 0
src/main/scala/com/aliyun/odps/spark/examples/makedata/makedata_02_writeredis.scala

@@ -35,6 +35,7 @@ object makedata_02_writeredis {
     val partition = partitionPrefix + date
     val savePathUser = param.getOrDefault("savePathUser", "")
     val savePathVideo = param.getOrDefault("savePathVideo", "")
+    val userSampleIDs = param.getOrDefault("userSampleIDs", "").split(",").toList
     // userDataTakeRddRun : /dw/recommend/model/
 
 
@@ -52,6 +53,9 @@ object makedata_02_writeredis {
     if (ifUser){
       println("user特征处理")
       var userData = odpsOps.readTable(project = project, table = tableUser, partition = partition, transfer = handleUser, numPartition = tablePart)
+      if (userSampleIDs.nonEmpty){
+        userData = userData.filter(r => userSampleIDs.contains(r._1.hashCode % 10))
+      }
       if (ifDebug){
         println("user特征处理-debug开启-只保留5条数据-特征数量大于1")
         val userDataTake = userData.filter(_._3 > 1).take(5)