add log

eda/node2vec/src/main/scala/com/gmei/Main.scala

@@ -127,13 +127,16 @@ object Main {
         .map { case (nodeId, (vector, name)) => (name,vector) }
         .repartition(200)
 
-
+      println("get node2vec")
+      println(node2vec_2.first())
 
       //2. compute similar cid and then take top k
 
       val storageLevel = StorageLevel.MEMORY_AND_DISK
 
       val indexed = node2vec_2.zipWithIndex.persist(storageLevel)
+      println("get indexed")
+      println(indexed.first())
 
       // create indexed row matrix where every row represents one word
       val rows = indexed.map {
@@ -141,11 +144,16 @@ object Main {
           IndexedRow(index, Vectors.dense(features))
       }
 
+      println("get index2vec")
+      println(rows.first())
+
       // store index for later re-mapping (index to word)
       val index = indexed.map {
         case ((word, features), index) =>
           (index, word)
       }.persist(storageLevel)
+      println("get index2node")
+      println(index.first())
 
       // create an input matrix from all rows and run lsh on it
       val matrix = new IndexedRowMatrix(rows)
@@ -197,7 +205,7 @@ object Main {
            |select a.device_id device_id,b.similarity_cid similarity_cid from
            |(select device_id,first(cid) as cid from data_feed_click
            |where cid_type='diary'
-           |and stat_date > '2018-09-25'
+           |and stat_date > '${date8}'
            |group by device_id) a left join
            |nd_cid_similarity_matrix b
            |on a.cid = b.cid