I have the following list containing tuples that have to values:
mylist=[(3, 'a'), (2, 'b'), (4, 'a'), (5, 'c'), (2, 'a'), (1, 'b')]
Is there a way to sum all values that share the same name? Something like:
(9, 'a'), (3, 'b'), (5, 'c')
I tried iterating tuples with for loop but can't get what i want.
Thank you
You can use itertools.groupby (after sorting by the second value of each tuple) to create groups. Then for each group, sum the first element in each tuple, then create a tuple per group in a list comprehension.
>>> import itertools
>>> [(sum(i[0] for i in group), key) for key, group in itertools.groupby(sorted(mylist, key = lambda i: i[1]), lambda i: i[1])]
[(9, 'a'), (3, 'b'), (5, 'c')]
Use a dict (or defaultdict) to aggregate over your tuples:
from collections import defaultdict
mylist=[(3, 'a'), (2, 'b'), (4, 'a'), (5, 'c'), (2, 'a'), (1, 'b')]
sums = defaultdict(int)
for i, k in mylist:
sums[k] += i
print sums.items()
# output: [('a', 9), ('c', 5), ('b', 3)]
# Or if you want the key/value order reversed and sorted by key
print [(v, k) for (k, v) in sorted(sums.items())]
# output: [(9, 'a'), (3, 'b'), (5, 'c')]
A simple approach without any dependencies:
mylist = [(3, 'a'), (2, 'b'), (4, 'a'), (5, 'c'), (2, 'a'), (1, 'b')]
sums = {}
for a, b in mylist:
if b not in sums: sums[b] = a
else: sums[b] += a
print list(sums.items())
You can create a dict which has sum of individual keys.
>>> my_dict = {}
>>> for item in mylist:
... my_dict[item[1]] = my_dict.get(item[1], 0) + item[0]
>>> [(my_dict[k], k) for k in my_dict]
[(9, 'a'), (5, 'c'), (3, 'b')]
