DataFrame.fillna

DataFrame.fillna(value=None, method=None, axis=None, inplace=False, limit=None, downcast=None)
使用指定的方法填充NA/NaN值。

Parameters

value 用来填充空的值
method ‘backfill’，’bfill’，’pad’，’ffill’
axis 0 or index：填充行，1 or columns：填充列
inplace False：返回一个副本；True：就地执行操作并返回None
limit 要填充空白的最大数目
downcast 表示如果可能要向下转换的类型

Example：value

  1. import pandas as pd
  3. df = pd.DataFrame({'site':['google', 'baidu', 'wiki'],
  4.                     'age':[18, 39, 22],
  5.                     'price': [None, 2.0, 3.0],
  6.                     'color': [None, 'black', None]})
  7. df.fillna(0)
  8. ----------------------------------------------
  9.     site    age    price    color
  10. 0    google    18    0.0        0
  11. 1    baidu    39    2.0        black
  12. 2    wiki    22    3.0        0
  14. df.fillna(value={'price': 1.0, 'color': 'red'})
  15. ----------------------------------------------
  16.     site    age    price    color
  17. 0    google    18    1.0        red
  18. 1    baidu    39    2.0        black
  19. 2    wiki    22    3.0        red

Example：method

  1. import pandas as pd
  3. df = pd.DataFrame({'site':['google', 'baidu', 'wiki'],
  4.                     'age':[18, 39, 22],
  5.                     'price': [None, 2.0, 3.0],
  6.                     'color': [None, 'black', None]})
  7.     site    age    price    color
  8. 0    google    18    NaN        None
  9. 1    baidu    39    2.0        black
  10. 2    wiki    22    3.0        None
  12. df.fillna(method="backfill")
  13. -----------------------------------------------
  14.     site    age    price    color
  15. 0    google    18    2.0        black
  16. 1    baidu    39    2.0        black
  17. 2    wiki    22    3.0        None
  19. df.fillna(method="ffill")
  20. -----------------------------------------------
  21.     site    age    price    color
  22. 0    google    18    NaN        None
  23. 1    baidu    39    2.0        black
  24. 2    wiki    22    3.0        black
  26. df.fillna(method="bfill")
  27. -----------------------------------------------
  28.     site    age    price    color
  29. 0    google    18    2.0        black
  30. 1    baidu    39    2.0        black
  31. 2    wiki    22    3.0        None
  33. df.fillna(method="pad")
  34. -----------------------------------------------
  35. site    age    price    color
  36. 0    google    18    NaN    None
  37. 1    baidu    39    2.0    black
  38. 2    wiki    22    3.0    black

Example：limit

  1. import pandas as pd
  3. df = pd.DataFrame({'site':['google', 'baidu', 'wiki'],
  4.                     'age':[18, 39, 22],
  5.                     'price': [None, 2.0, 3.0],
  6.                     'color': [None, 'black', None]})
  7. df.fillna(value={'price': 1.0, 'color': 'red'}, limit=1)
  8. ----------------------------------------------------------
  9.     site    age    price    color
  10. 0    google    18    1.0        red
  11. 1    baidu    39    2.0        black
  12. 2    wiki    22    3.0        None