khuyentran1401/spark.py Secret

## spark.py
# schema: *
def get_median(df: pd.DataFrame) -> List[Dict[str, Any]]:
    return [{"col1": df["col1"].median(), "col2": df["col2"].iloc[0]}]

## spark.sql
%%fsql spark
SELECT *
FROM df
TRANSFORM PREPARTITION BY col2 USING get_median
PRINT
	# schema: *
	def get_median(df: pd.DataFrame) -> List[Dict[str, Any]]:
	return [{"col1": df["col1"].median(), "col2": df["col2"].iloc[0]}]
	%%fsql spark
	SELECT *
	FROM df
	TRANSFORM PREPARTITION BY col2 USING get_median
	PRINT