如果某个键相同，则获取所有值的平均值

Question

So I have this list of dictionaries:

l = [{'COUNTRY': 'UK', 'CREDITS': '54'}, {'COUNTRY': 'PT', 'CREDITS': '100'}, {'COUNTRY': 'FR', 'CREDITS': '20'}, {'COUNTRY': 'UK', 'CREDITS': '30'}, {'COUNTRY': 'UK', 'CREDITS': '15'}, {'COUNTRY': 'PT', 'CREDITS': '35'}, {'COUNTRY': 'FR', 'CREDITS': '30'}]

I need to get the average of the credits of each country. The output should be something like this:

l2 = {'UK': '102', 'PT': '67.5', 'FR': '25'}

Is there any simple and easy way to implement this?

Answer 1

I would create a defaultdict first to gather the values in a list of integers under the "COUNTRY" key.

Then I'll create a dict comprehension, performing the mean:

l = [{'COUNTRY': 'UK', 'CREDITS': '54'}, {'COUNTRY': 'PT', 'CREDITS': '100'}, {'COUNTRY': 'FR', 'CREDITS': '20'}, {'COUNTRY': 'UK', 'CREDITS': '30'}, {'COUNTRY': 'UK', 'CREDITS': '15'},
 {'COUNTRY': 'PT', 'CREDITS': '35'}, {'COUNTRY': 'FR', 'CREDITS': '30'}]

import collections

d = collections.defaultdict(list)
for s in l:
    d[s["COUNTRY"]].append(int(s["CREDITS"]))

result = {k:sum(v)/len(v) for k,v in d.items()}

print(result)

result:

{'UK': 33.0, 'PT': 67.5, 'FR': 25.0}

note that 1) your expected result is wrong and 2) I converted to float, but you can leave it as integer as string by doing

result = {k:str(sum(v)//len(v)) for k,v in d.items()}

which gives:

{'PT': '67', 'FR': '25', 'UK': '33'}

Answer 2

Alternative solution using itertools.groupby() and itertools.tee() functions:

import itertools

l = [{'COUNTRY': 'UK', 'CREDITS': '54'}, {'COUNTRY': 'PT', 'CREDITS': '100'}, {'COUNTRY': 'FR', 'CREDITS': '20'}, {'COUNTRY': 'UK', 'CREDITS': '30'}, {'COUNTRY': 'UK', 'CREDITS': '15'}, {'COUNTRY': 'PT', 'CREDITS': '35'}, {'COUNTRY': 'FR', 'CREDITS': '30'}]
avgs = {}

for k,g in itertools.groupby(sorted(l, key=lambda x: x['COUNTRY']), key=lambda x: x['COUNTRY']):
    d1,d2 = itertools.tee(g)  # copy `grouper` iterator to deal with "fresh" pointer
    avgs[k] = sum(int(d['CREDITS']) for d in d1)/len(list(d2))

print(avgs)

The output:

{'UK': 33.0, 'FR': 25.0, 'PT': 67.5}