python - 用 np.NaN 替换 pandas 数据框中的缺失值(以字符串形式给出)

Question

我有一个数据框energy，其中某些列缺少值。缺失值由数据框中的字符串 ... 表示。我想用 np.NaN

替换所有这些值

In [3]: import pandas as pd

In [4]: import numpy as np

In [7]: energy = pd.read_excel('test.xls', skiprows = 17, skip_footer = 38, parse_cols = range(2, 6), index_col = None, names = ['Country', 'ES'
   ...: , 'ESC', '% Renewable'])

In [8]: energy[(energy['ES'] == "...") | (energy['ESC'] == "...")]
Out[8]: 
                          Country   ES  ESC  % Renewable
3                  American Samoa  ...  ...     0.641026
86                           Guam  ...  ...     0.000000
150      Northern Mariana Islands  ...  ...     0.000000
210                        Tuvalu  ...  ...     0.000000
217  United States Virgin Islands  ...  ...     0.000000

为了替换这些值，我尝试过:

In [9]: energy[(energy['ES'] == "...")]['ES'] = np.NaN
/usr/local/bin/ipython:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
  #!/usr/bin/python3

我不明白这个错误，而且我也没有看到任何其他方法来实现我想要的。有什么想法吗？

Answer 1

我认为你需要:

energy['ES'] = energy.loc[energy['ES'] != "...", 'ES'] 

另一个解决方案:

energy['ES'] = energy['ES'].mask(energy['ES'] == "...")

或者:

energy['ES'] = energy['ES'].replace({'...': np.nan})

但最好的是 ayhan 评论:

you can pass na_values='...' to pd.read_excel