词典列表中键值的平均值

Question

I have this list of dictionaries :

[{'Eva': [4, 8, 2]}, {'Ana': [57, 45, 57]}, {'Ada': [12]}]

I need to get the average of each key value, so that the output is :

[{'Eva': [5], {'Ana' : [53]}, {'Ada':[12]}]

The average must be rounded up or down by adding 0.5 and taking the only the integer part. For example, if the average is 4.3 adding 0.5 equals 4.8, so the output is 4. If the average is 4.6 adding 0.5 equals 5.1, so the output is 5.

I know how to use iteritems() to iterate over a dictionary, but since this is a list I don't know how to reach every value.

Thanks in advance.

Answer 1

Define an average function and then use a nested list / dict comprehension to get the averages for each key for each dictionary in the list.

>>> from __future__ import division
>>> lst = [{'Eva': [4, 8, 2]}, {'Ana': [57, 45, 57]}, {'Ada': [12]}]
>>> avg = lambda x: int(round(sum(x) / len(x)))
>>> [{k: avg(v) for k, v in d.items()} for d in lst]
[{'Eva': 5}, {'Ana': 53}, {'Ada': 12}]

If you want the average to be wrapped in a list, just use [avg(v)] :

>>> [{k: [avg(v)] for k, v in d.items()} for d in lst]
[{'Eva': [5]}, {'Ana': [53]}, {'Ada': [12]}]

Answer 2

You can use zip and numpy functions mean and round for this task:

In [8]: import numpy as np

In [9]:  [dict(zip(d.keys(), [int(np.round(np.mean(d.values())))])) for d in L]

#Out[9]: [{'Eva': 5}, {'Ana': 53}, {'Ada': 12}]

Version with "less" parenthesis:

[dict(zip(d.keys(), [np.array(d.values()).mean().round().astype(int)])) for d in L]

Answer 3

This is an "easy to read and understand" solution. I guess from your question that you do not have too much confidence with python so, I will keep it very basic.

myList = [{'Eva': [4, 8, 2]}, {'Ana': [57, 45, 57]}, {'Ada': [12]}]

result = []
for item in myList:
    for k,v in item.items(): #use iteritems() if Python2
        average = round(sum(v)/len(v)) #round to the closest integer
        result.append({k:average})

>>>print(result)
[{'Eva': 5}, {'Ana': 53}, {'Ada': 12}]

Answer 4

You can do:

data = [{'Eva': [4, 8, 2]}, {'Ana': [57, 45, 57]}, {'Ada': [12]}]
outdata = []

for item in data:
    tmp = {}
    for key, val in item.iteritems():
        avg = reduce(lambda x, y: x + y, val) / len(val)
        tmp[key] = [avg]
    outdata.append(tmp)

Which will also work in the case where you have dictionaries with multiple keys (although the additional variable is probably wasteful if that can never happen).

Answer 5

Oneliner:

In:

[{el.keys()[0]: [int( round(1.0 * sum(el.values()[0]) / len(el.values()[0]) ))]} for el in li]

Out:

[{'Eva': [5]}, {'Ana': [53]}, {'Ada': [12]}]