Python 将多列拆分为多列，Pandas

对于矢量版本，您可以临时使用stack和str.split:

df.stack().str.split().unstack()

输出:

             ≤8          9–13       14–15       16–17       18–20         ≥21
1  [3687, 55.5]  [2234, 33.6]  [290, 4.4]  [194, 2.9]  [185, 2.8]   [52, 0.8]
2   [838, 66.5]   [419, 33.3]         NaN         NaN         NaN    [0, 0.0]
3  [8905, 66.9]  [3362, 25.2]  [473, 3.6]  [252, 1.9]  [184, 1.4]  [144, 1.1]

要使多个列使用expand=True参数，您将得到一个多索引，然后可以根据需要进行修改:

df.stack().str.split(expand=True).unstack()

输出:

      0                                  1                             
     ≤8  9–13 14–15 16–17 18–20  ≥21    ≤8  9–13 14–15 16–17 18–20  ≥21
1  3687  2234   290   194   185   52  55.5  33.6   4.4   2.9   2.8  0.8
2   838   419   NaN   NaN   NaN    0  66.5  33.3   NaN   NaN   NaN  0.0
3  8905  3362   473   252   184  144  66.9  25.2   3.6   1.9   1.4  1.1

reorganizing the order

(df.stack()
   .str.split(expand=True)
   .unstack()
   .swaplevel(axis=1)
   [df.columns]
)

输出:

     ≤8        9–13       14–15      16–17      18–20       ≥21     
      0     1     0     1     0    1     0    1     0    1    0    1
1  3687  55.5  2234  33.6   290  4.4   194  2.9   185  2.8   52  0.8
2   838  66.5   419  33.3   NaN  NaN   NaN  NaN   NaN  NaN    0  0.0
3  8905  66.9  3362  25.2   473  3.6   252  1.9   184  1.4  144  1.1