import org.apache.spark.sql.expressions.Window
val partitionBy = Window.partitionBy("name", "sit").orderBy("data_date")
val newDf = df.withColumn("PreviousDate", lag("uniq_im",
1).over(partitionBy))
Cheers...
On Thu, Mar 14, 2019 at 4:55 AM anbu wrote:
> Hi,
>
> To calculate LAG functions
Hi,
To calculate LAG functions difference for the two data_date(current date and
previous date) on the same column
Could you please help me to implement the below scenario using scala spark
Dataset.
uniq_im - LAG(uniq_im,1,0) OVER PARTITION BY(name,sit,plc,country,state)
order by (data_date) as
