Aan de slagGa gratis aan de slag

Spark SQL Join

Sometimes it is much easier to write complex joins in SQL. In this exercise, we will start with the join keys already in the same format and precision but will use SparkSQL to do the joining.

Deze oefening maakt deel uit van de cursus

Feature Engineering with PySpark

Cursus bekijken

Oefeninstructies

  • Register the Dataframes as SparkSQL tables with createOrReplaceTempView, name them the df and walk_df respectively.
  • In the join_sql string, set the left table to df and the right table to walk_df
  • Call spark.sql() on the join_sql string to perform the join.

Praktische interactieve oefening

Probeer deze oefening eens door deze voorbeeldcode in te vullen.

# Register dataframes as tables
____.createOrReplaceTempView(____)
____.createOrReplaceTempView(____)

# SQL to join dataframes
join_sql = 	"""
			SELECT 
				*
			FROM ____
			LEFT JOIN ____
			ON df.longitude = walk_df.longitude
			AND df.latitude = walk_df.latitude
			"""
# Perform sql join
joined_df = spark.sql(____)
Code bewerken en uitvoeren