Sorting dictionary - functools.reduce ignores negative values?

Question

I have a dictionary like below:

d = [{152: 1}, {151: 3}, {152: 4}, {153: 5}, {153: 10}, {154: -0.1}]

I want to sort them like below:

{151: 3, 152: 5, 153: 15, 154: -0.1}

I tried the below by following this answer ...

sorted = dict(functools.reduce(operator.add, map(collections.Counter, d)))

The output I got is :

{151: 3, 152: 5, 153: 15}

I do not understand why functools.reduce(operator.add, ) removes the {154: -0.1}. Isn't operator.add just a plain summation method?

operator.add(a, b)
operator.add(a, b)
Return a + b, for a and b numbers.

Questions

What could be the workaround to achieve the desired output?
Why does functools.reduce(operator.add, ) remove the negative value?

d is not a dict but a lit of dicts – hpchavaz Dec 05 '21 at 13:44 — hpchavaz, Dec 05 '21 at 13:44

Alain T. · Answer 1 · 2021-12-05T14:06:34.033

1

The Counter class is designed for positive counts so -0.1 is treated as 0 by its addition method (as would any other negative result). You could do this using a loop and a regular dictionary (no library needed):

dicts = [{152: 1}, {151: 3}, {152: 4}, {153: 5}, {153: 10}, {154: -0.1}]

r = dict()
for d in dicts:
    r.update({k:r.get(k,0)+v for k,v in d.items()})

print(r)
{152: 5, 151: 3, 153: 15, 154: -0.1}

The documentation for Counter states that Each operation can accept inputs with signed counts, but the output will exclude results with counts of zero or less

edited Dec 05 '21 at 14:06

answered Dec 05 '21 at 13:54

Alain T.

40,517
4
31
51

And to clarify from OP's question: "Why does functools.reduce(operator.add, ) remove the negative value?" It's not that functools.reduce(operator.add, ...) "removes" the negative value; rather, when one Counter object is added to another Counter object (in this case via `functools.reduce`), the resulting Counter object will not contain any key, value pairs where the value would be less than or equal to 0. – Farhaan S. Dec 05 '21 at 14:42

score 0 · Answer 2 · answered Dec 05 '21 at 15:17

0

A pandas solution

>>>import pandas as pd
>>>d = [{152: 1}, {151: 3}, {152: 4}, {153: 5}, {153: 10}, {154: -0.1}]
>>>pd.DataFrame(d).sum().to_dict()
{152: 5.0, 151: 3.0, 153: 15.0, 154: -0.1}

answered Dec 05 '21 at 15:17

hpchavaz

1,368
10
16

score 0 · Answer 3 · answered Dec 05 '21 at 16:21

operator.add is a plain summation method, please see its source code: operator.add

Why does functools.reduce(operator.add, ) remove the negative value?

Your one-line code is equivalent to Counter({152: 1}) + Counter({151: 3}) + Counter({152: 4}) + Counter({153: 5}) + Counter({153: 10}) + Counter({154: -0.1})

However, the add method for Counter will remove negative values, check its source: counter-add

Thus, the value for key 154 will be ignored.

How to make your work done?

t = {}
for i in d:
    for k in i.keys():
        t[k] = t[k]+i[k] if k in t else i[k]

Sorting dictionary - functools.reduce ignores negative values?

3 Answers3