When using pivot, the output column names are different in PySpark and teradatamlspk.
PySpark
>>> df.groupBy("year").pivot("course", ["dotNET", "Java"]).sum("earnings").show()
+----+------+-----+ |year|dotNET| Java| +----+------+-----+ |2012| 15000|20000| |2013| 48000|30000| +----+------+-----+
teradatamlspk
>>> df.groupBy("year").pivot("course", ["dotNET", "Java"]).sum("earnings").show()
+----+-------------------+-----------------+ |year|sum_earnings_dotnet|sum_earnings_java| +----+-------------------+-----------------+ |2012| 15000| 20000| |2013| 48000| 30000| +----+-------------------+-----------------+
When using pivot in teradatamlspk, the grouping columns are not returned.
PySpark
>>> df1.groupBy("year").pivot("course", ["dotNET", "Java"]).sum("year").show()
+----+------+----+ |year|dotNET|Java| +----+------+----+ |2012| 4024|2012| |2013| 2013|2013| +----+------+----+
teradatamlspk
>>> df1.groupBy("year").pivot("course", ["dotNET", "Java"]).sum("year").show()
+---------------+-------------+ |sum_year_dotnet|sum_year_java| +---------------+-------------+ | 6037| 4025| +---------------+-------------+