从 Pandas 数据框中提取 Frozenset 项目

Question

我有以下数据框：

我想将 "antecedents" 和 "consequents" 列转换为字符串，删除 "frozenset({ ... })" 格式，因此对于所有行：

"VENTOLIN S.INAL200D 100MCG"，而不是 frozenset({ "VENTOLIN S.INAL200D 100MCG" }).

我设法实现了结果：

prod = []

for i in df["antecedents"]:
    prod.append(str(i))

new_set = {x.replace('frozenset', ''
                     ).replace('})', ''
                        ).replace('({', ''
                        ).replace("'", "") for x in prod}

是否有更pythonic的解决方案？

Answer 1

首先将值转换为元组或列表，然后使用 DataFrame.explode:

df = pd.DataFrame({
         'antecedents':[frozenset({'aaa', 'bbb'})] * 3 + [frozenset({'nbb'})] * 3,
         'consequents':[frozenset({'ccc'})] * 3 + [frozenset({'nbb', 'ddd'})] * 3,
         'C':[1,3,5,7,1,0],
})
#print (df)

cols = ['antecedents','consequents']
df[cols] = df[cols].applymap(lambda x: tuple(x))
print (df)
  antecedents consequents  C
0  (bbb, aaa)      (ccc,)  1
1  (bbb, aaa)      (ccc,)  3
2  (bbb, aaa)      (ccc,)  5
3      (nbb,)  (nbb, ddd)  7
4      (nbb,)  (nbb, ddd)  1
5      (nbb,)  (nbb, ddd)  0

df1 = (df.explode('antecedents')
         .reset_index(drop=True)
         .explode('consequents')
         .reset_index(drop=True))
print (df1)
   antecedents consequents  C
0          bbb         ccc  1
1          aaa         ccc  1
2          bbb         ccc  3
3          aaa         ccc  3
4          bbb         ccc  5
5          aaa         ccc  5
6          nbb         nbb  7
7          nbb         ddd  7
8          nbb         nbb  1
9          nbb         ddd  1
10         nbb         nbb  0
11         nbb         ddd  0

从 Pandas 数据框中提取 Frozenset 项目

Extract Frozenset items from Pandas Dataframe

python

pandas

frozenset