Odtnhj

Write a function named sort_by_average_rating that takes a list/array of key-value stores as a parameter where each key-value store has keys "ratings", "budget", and "box_office" where "budget" and "box_office" are integers and "ratings" is a list of integers.

Sort the input based on the average of the values in "ratings".

def sort_by_average_rating(lista):
 lista.sort(sum(lista['ratings']) / int(len(lista['ratings'])))
 return lista

I am getting error on input:

error on input ['ratings': [9, 10, 2, 3, 8, 10, 9], 'budget': 16219606.11, 'box_office': 13297812, 'ratings': [6, 3, 1, 7, 9, 2], 'budget': 12995254.35, 'box_office': 26409035] list indices must be integers or slices, not str.

Expected Output:

['ratings': [6, 3, 1, 7, 9, 2], 'budget': 12995254.35, 'box_office': 26409035, 'ratings': [9, 10, 2, 3, 8, 10, 9], 'budget': 16219606.11, 'box_office': 13297812]

What am I doing wrong?

Your error message says you have this list of dictionaries:

L = ['ratings': [9, 10, 2, 3, 8, 10, 9], 'budget': 16219606.11, 'box_office': 13297812,
 'ratings': [6, 3, 1, 7, 9, 2], 'budget': 12995254.35, 'box_office': 26409035]

You can just use sorted with a custom key function:

res = sorted(L, key=lambda x: sum(x['ratings']) / len(x['ratings']))

['box_office': 26409035, 'budget': 12995254.35, 'ratings': [6, 3, 1, 7, 9, 2],
 'box_office': 13297812, 'budget': 16219606.11, 'ratings': [9, 10, 2, 3, 8, 10, 9]]

Alternatively, you can use statistics.mean:

from statistics import mean

res = sorted(L, key=lambda x: mean(x['ratings']))

Notice the lambda function takes each item in your iterable (L) rather than the entire list of dictionaries (as in your current attempt).