python - 重复行以获取所有星期的数据

Question

我想知道是否有一种不太笨拙的方法来为以下示例创建所有星期的所有组合:

df = pd.DataFrame({
    'Gender': ['female', 'female', 'female', 'female'],
    'Bla': ['a', 'a', 'a', 'a'],
    'Week': [1, 2, 3, 4]
})
print(df)

number_of_weeks = 52
temp = df[['Gender', 'Bla']].drop_duplicates()
results = temp.loc[temp.index.repeat(number_of_weeks)]
results["Week"] = range(1, len(results) + 1, 1)
print(results)

这种工作，但它也应该适用于分类变量超过 1 个“级别”的情况。

df = pd.DataFrame({
    'Gender': ['female', 'female', 'female', 'female'],
    'Bla': ['a', 'b', 'a', 'a'],
    'Week': [1, 2, 3, 4]
})

上面的代码会将其扩展到不可行的周数。

Answer 1

尝试使用concat:

temp = df[["Gender", "Bla"]].drop_duplicates()
output = pd.concat([temp]*52, ignore_index=True).assign(week=list(range(1,53))*temp.shape[0])

>>> output
     Gender Bla  week
0    female   a     1
1    female   b     2
2    female   a     3
3    female   b     4
4    female   a     5
..      ...  ..   ...
99   female   b    48
100  female   a    49
101  female   b    50
102  female   a    51
103  female   b    52

[104 rows x 3 columns]