final code

main.py

@@ -1,4 +1,3 @@
-from pyspark.sql import SparkSession
 from pyspark.sql.functions import col
 import argparse
 
@@ -8,17 +7,27 @@ def transform_data(data_source: str, output_uri: str) -> None:
 
         df.createOrReplaceTempView("bigbasket_products")
 
-        SQL_QUERY = """
-            select product, count(*) 
+        SQL_QUERY = ["""
+            select avg(market_price) - avg(sale_price)
             from bigbasket_products
-            group by product
+            """,
                 """
-
-        transformed_df = spark.sql(SQL_QUERY)
-
-        print(f"Number of rows: {transformed_df.count()}")
-
-        transform_df.write.mode("overwrite").parquet(output_uri)
+            select avg(rating)
+            from bigbasket_products
+                """,
+                """
+            select avg(market_price)
+            from bigbasket_products
+                """,
+                """
+            select avg(sale_price)
+            from bigbasket_products
+                """
+        ]
+        for i in SQL_QUERY:
+            transformed_df = spark.sql(i)
+            transformed_df.show()
+            transformed_df.write.mode("overwrite").parquet(output_uri)
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()