How can I compare Multiple lists of tuples like this:
[[(1,2), (3,6), (5,3)], [(1,5), (3,5)], [(2,1), (1,8), (3,9)]]
The output should be:
[(1,2), (1,5), (1,8)],[(3,6), (3,5), (3,9)]
It means that i want just those values whose x-axis value matches others.
(5,3) and (2,1) should be discarded!
解决方案
One possible Option
>>> def group(seq):
for k, v in groupby(sorted(chain(*seq), key = itemgetter(0)), itemgetter(0)):
v = list(v)
if len(v) > 1:
yield v
>>> list(group(some_list))
[[(1, 2), (1, 5), (1, 8)], [(3, 6), (3, 5), (3, 9)]]
Another Popular Option
>>> from collections import defaultdict
>>> def group(seq):
some_dict = defaultdict(list)
for e in chain(*seq):
some_dict[e[0]].append(e)
return (v for v in some_dict.values() if len(v) > 1)
>>> list(group(some_list))
[[(1, 2), (1, 5), (1, 8)], [(3, 6), (3, 5), (3, 9)]]
So which of them fairs better with the example data?
>>> def group_sort(seq):
for k, v in groupby(sorted(chain(*seq), key = itemgetter(0)), itemgetter(0)):
v = list(v)
if len(v) > 1:
yield v
>>> def group_hash(seq):
some_dict = defaultdict(list)
for e in chain(*seq):
some_dict[e[0]].append(e)
return (v for v in some_dict.values() if len(v) > 1)
>>> t1_sort = Timer(stmt="list(group_sort(some_list))", setup = "from __main__ import some_list, group_sort, chain, groupby")
>>> t1_hash = Timer(stmt="list(group_hash(some_list))", setup = "from __main__ import some_list, group_hash,chain, defaultdict")
>>> t1_hash.timeit(100000)
3.340240917954361
>>> t1_sort.timeit(100000)
0.14324535970808938
And with a much larger random list
>>> some_list = [[sample(range(1000), 2) for _ in range(100)] for _ in range(100)]
>>> t1_sort.timeit(100)
1.3816694363194983
>>> t1_hash.timeit(1000)
34.015403087978484
>>>