Python随笔（i1）：DataFrame多条件“与”选择 Parallel Criteria Selection in DataFrame

可以使用&操作符连接由形为dataframe_obj[col] Op arg的语句所创建的布尔值Series。

import pandas


data = {
    "Name": [ "pencil", "ruler", "compass" ],
    "Price": [ 1.5, 4.3, 8.1 ],
    "InStock": [ False, True, True ]
}

df = pandas.DataFrame(data)

print("Mantle has 5 yuan so he can afford the following stuff:")
print(df[df["Price"] < 5.0])
print("... but he can only buy things in stock so he can actually get:")
print(df[(df["Price"] <= 5.0) & (df["InStock"] == True)])

结果：

mantlebao@mantletmx:~# python3 pandas_comb_cond.py       
Mantle has 5 yuan so he can afford the following stuff:
     Name  Price  InStock
0  pencil    1.5    False
1   ruler    4.3     True
... but he can only buy things in stock so he can actually get:
    Name  Price  InStock
1  ruler    4.3     True
mantlebao@mantletmx:~#

同样也可以使用dataframe_obj.where(crit, inplace=False)方法，该函数会将所有不符合要求的行的所有字段设为NaN。