python Pandas : change only next unnamed column with the previous column name

Question

是否有任何方法可以只为下一列提供下一列和上一列的未命名列。

输入:

ca~                         ch~     
Campaign Type   Code    Channel Type    Code    Channel Type Definition

输出:

    ca~             ca~     ch~             ch~
    Campaign Type   Code    Channel Type    Code    Channel Type Definition

我试过下面但它填满了所有的空白，虽然我只想填写下一列而不是另一个空白。

x.columns = x.columns.to_series().mask(lambda y: y.str.startswith('Unnamed')).ffill()

Answer 1

示例

             ca~ Unnamed           ch~ Unnamed.1                Unnamed.2
0  Campaign Type    Code  Channel Type      Code  Channel Type Definition

一种方法

df.columns = [df.columns[i-1] if 'Unnamed' in df.columns[i] else df.columns[i] for i in range(len(df.columns))]

---

             ca~   ca~           ch~   ch~                Unnamed.1
0  Campaign Type  Code  Channel Type  Code  Channel Type Definition

备选

先用NaN替换'Unnamed'然后用limit = 1 ffill

df.columns = list(pd.DataFrame(df.columns).replace('Unnamed\.?\d?',np.nan,regex=True).ffill(limit=1)[0])

---

    ca~   ca~           ch~   ch~                      NaN
0  Campaign Type  Code  Channel Type  Code  Channel Type Definition