spark-get-max-row.scala

scala> val values = Seq((1, "a",1), (1, "b",2), (2, "c", 2), (3, "d",1), (3, "e", 1), (3, "f",0))
values: Seq[(Int, String, Int)] = List((1,a,1), (1,b,2), (2,c,2), (3,d,1), (3,e,1), (3,f,0))

scala> val df = values.toDF
df: org.apache.spark.sql.DataFrame = [_1: int, _2: string ... 1 more field]

scala> val max_df = df.groupBy("_1").agg(max("_3").alias("_3"))
max_df: org.apache.spark.sql.DataFrame = [_1: int, _3: int]

scala> df.join(max_df, Seq("_1", "_3"), "leftsemi").dropDuplicates("_1", "_3").show
+---+---+---+
| _1| _3| _2|
+---+---+---+
|  3|  1|  d|
|  2|  2|  c|
|  1|  2|  b|
+---+---+---+