python - How to get a total count based on distinction of two columns with PySpark?

Question

How do I do a summation of the frequency based on distinct ID & Location in PySpark?

Feels like I need to do window partition by ID and Location and then add the frequency but not sure how to write this in Pyspark code:

Input

ID	Location	Frequency
AAA	Mcd	2
AAA	Mcd	1
BBB	Nandos	1
BBB	Nandos	3
AAA	KFC	2
BBB	KFC	4

question from:https://stackoverflow.com/questions/66058530/how-to-get-a-total-count-based-on-distinction-of-two-columns-with-pyspark

Answer 1

Just a simple group by and sum:

import pyspark.sql.functions as F

df2 = df.groupBy('ID', 'Location').agg(F.sum('Frequency').alias('TotalFrequency'))