[Python] Counter & combinations & set

---

Writing Efficient Python Code

• PART3
• Counter from module collections
  • 종류별 개수를 반환
• combinations from itertools
  • 가능한 조합을 list로 반환
• set from itertools
  • set을 통해 distinct한 element만 산출
  • intersection을 통해 교집합 산출
  • difference을 통해 차집합 산출
  • symmetric_difference을 통해 합집합-교집합 산출
• 계산이 한 번만 필요한 것은 loop 전에 계산하기
• map을 활용해 한 번에 처리하기

---

Counter

1) Counter

import seaborn as sns
import pandas as pd
from collections import Counter

iris_df = sns.load_dataset('iris')
Counter(iris_df['species'])

# 위와 같음
# species_dict = {}
# for species in iris_species:
#     print(species
#     if species in species_dict.keys():
#         species_dict[species] += 1
#     else:
#         species_dict[species] = 1

Counter({'setosa': 50, 'versicolor': 50, 'virginica': 50})

combinations

2) combinations

from itertools import combinations

iris_species = list(set(iris_df['species']))

# get every combination of 2
print(list(combinations(iris_species, 2)))

[('setosa', 'virginica'), ('setosa', 'versicolor'), ('virginica', 'versicolor')] ---

set

3) intersection & difference & symmetric_difference

set_a = set([1, 2, 3])
set_b = set([3, 4, 5])

# 교집합
both = set_a.intersection(set_b)
print(both)

# 차집합
set_a_only = set_a.difference(set_b)
print(set_a_only)

# 합집합-교집합
unique_to_set = set_a.symmetric_difference(set_b)
print(unique_to_set)

{3}
{1, 2}
{1, 2, 4, 5} ---

Better loops

# combinations()
possible_pairs = [*combinations(set(iris_df['species']),2)]
enumerated_tuples = []

for i,pair in enumerate(possible_pairs, 1):
    enumerated_pair_tuple = (i,) + pair
    enumerated_tuples.append(enumerated_pair_tuple)

enumerated_pairs = [*map(list, enumerated_tuples)]
print(enumerated_pairs)


# 위와 같음
# for i,pair in enumerate(possible_pairs, 1):
#     enumerated_pair_tuple = (i,) + pair
#     enumerated_pair_list = list(enumerated_pair_tuple)
#     enumerated_pairs.append(enumerated_pair_list)

[[1, 'setosa', 'virginica'], [2, 'setosa', 'versicolor'], [3, 'virginica', 'versicolor']]

---