Merge branch 'master' of git.wanmeizhensuo.com:ML/ffm-baseline

增加编码格式

eda/feededa/src/main/scala/com/gmei/EsmmData.scala

@@ -284,11 +284,20 @@ object EsmmPredData {
 
       val raw_data2 = sc.sql(
         s"""
-           |select device_id,city_id,explode(split(merge_queue, ',')) as cid_id from raw_data1
+           |select device_id,city_id,merge_queue from raw_data1 limit 10000
          """.stripMargin
       )
-      raw_data2.createOrReplaceTempView("raw_data")
-      print(raw_data2.count())
+      raw_data1.createOrReplaceTempView("raw_data2")
+      println(raw_data2.count())
+      raw_data2.show()
+
+      val raw_data3 = sc.sql(
+        s"""
+           |select device_id,city_id as ucity_id,explode(split(merge_queue, ',')) as cid_id from raw_data2
+         """.stripMargin
+      )
+      raw_data3.createOrReplaceTempView("raw_data")
+      println(raw_data3.count())
 
 
 
@@ -310,7 +319,8 @@ object EsmmPredData {
            |where b.partition_date = '${yesteday}'
          """.stripMargin
       )
-      sid_data.show()
+//      sid_data.show()
+      println(sid_data.count())
 
       val sid_data_label = sid_data.withColumn("y",lit(0)).withColumn("z",lit(0))
       sid_data_label.createOrReplaceTempView("union_data")
@@ -372,7 +382,7 @@ object EsmmPredData {
          """.stripMargin
       )
       //      union_data_scity_id.createOrReplaceTempView("union_data_scity_id")
-      union_data_scity_id.show()
+      println(union_data_scity_id.count())
       GmeiConfig.writeToJDBCTable("jdbc:mysql://10.66.157.22:4000/jerry_test?user=root&password=3SYz54LS9#^9sBvC&rewriteBatchedStatements=true",union_data_scity_id, table="esmm_pre_data",SaveMode.Overwrite)
 
 

eda/feededa/src/main/scala/com/gmei/GmeiConfig.scala

@@ -33,7 +33,7 @@ object GmeiConfig extends Serializable {
   def getSparkSession():(SparkContext, SparkSession) = {
     val sparkConf = new SparkConf
     sparkConf.set("spark.sql.crossJoin.enabled", "true")
-    sparkConf.set("spark.debug.maxToStringFields", "100")
+    sparkConf.set("spark.debug.maxToStringFields", "130")
     sparkConf.set("spark.sql.broadcastTimeout",  "6000")
 
     if (!sparkConf.contains("spark.master")) {