How to pivot table in pyspark

df= spark.read.format("csv").option("header","True").load("file:///home/cloudera/product.csv")

df.show()

dff = df.groupBy("PRODUCT_KEY").pivot("YEAR").agg(first("volume"))

dff.show()