Multi-Join Query

In this example we use VerbaGPT to get a semi-complex query involving multiple joins using the Chinook dataset¹.

Video:

User question: what are the top 3 selling genre in chinook data?

VerbaGPT response:

Cost of query: $0.00

Name  TotalSold

Rock 835
Latin 386
Metal 264

VerbaGPT answers the question correctly. The query seems innocuous enough, but there is more going on under the surface.

Check out the code below, answering this question involved joins across three tables and VerbaGPT was able to leverage the power of gpt4 to get the question answered.

import pandas as pd

query = '''
SELECT TOP 3 g.Name, SUM(il.Quantity) AS TotalSold
FROM Genre g
JOIN Track t ON g.GenreId = t.GenreId
JOIN InvoiceLine il ON t.TrackId = il.TrackId
GROUP BY g.Name
ORDER BY TotalSold DESC
'''

df = pd.read_sql(query, conn)
print(df)

1. Chinook dataset