itertools.chain.from_iterable 适用于嵌套数字列表但不适用于字符串列表？

Question

我的数据框中有一列由列表组成，我想将每一行的所有列表组合成一个单元格中的一个列表。这是列的样子df.terms.dropna()0     [Algorithms, Brain, Brain Mapping, Computer Si...4     [Adult, Algorithms, Cerebrovascular Circulatio...5     [Algorithms, Brain, Brain Mapping, Hemodynamic...7     [Adult, Algorithms, Brain, Cerebrovascular Cir...10    [Animals, Base Composition, Birds, Genetic Var...Name: mesh_terms, dtype: object我设法将它们结合在一起并得到0    [[Algorithms, Brain, Brain Mapping, Computer S...],[Adult, Algorithms, Cerebrovascular Circulatio...],[Algorithms, Brain, Brain Mapping, Hemodynamic...],[list_index_7],[list_index_10]]Name: mesh_terms, dtype: object但我想要一个包含所有字符串的长列表，例如[Algorithms, Brain, Brain Mapping, Computer Si..., ... , Animals, Base Composition, Birds, Genetic Var...]我试过使用 itertools，但它仍然给了我一个嵌套列表，但它适用于这个例子list2d = [[1,2,3],[4,5,6], [7], [8,9]]list(itertools.chain.from_iterable(list2d))[1, 2, 3, 4, 5, 6, 7, 8, 9]也试过flattened = [val for sublist in list_of_lists for val in sublist]没有让它工作。

Answer 1

将值转换为列表，然后转换为DataFrame或Series构造函数：

df_mesh = pd.DataFrame({'terms': [['Algorithms','Brain'],['Adult','Algorithms']]})

print (df_mesh)

                 terms

0  [Algorithms, Brain]

1  [Adult, Algorithms]

df = pd.DataFrame({'new': [df_mesh['terms'].tolist()]})

print (df)

                                          new

0  [[Algorithms, Brain], [Adult, Algorithms]]

s = pd.Series([df_mesh['terms'].tolist()])

print (s)

0    [[Algorithms, Brain], [Adult, Algorithms]]

dtype: object

编辑：

s1 = pd.Series([[val for sublist in df_mesh['terms'] for val in sublist]])

print (s1)

0    [Algorithms, Brain, Adult, Algorithms]

dtype: object

或者：

s1 = pd.Series([list(itertools.chain.from_iterable(df_mesh['terms']))])