Question

我的数据框看起来像这样：

   Major        Sample_size   Men   Women  ShareWomen  Employed    ...     Part_time
   Economics    36            2057  282    0.120564    1976        ...     270   
   French       7             679   77     0.101852    640         ...     170

我试图按如下方式定义一个函数：

def cutoff(category, cut, direction):

    if direction == 0:
        comply = list(zip(df[df.category < cut].Major, df[df.category < cut].category))

    if direction == 1:
        comply = list(zip(df[df.category > cut].Major, df[df.category > cut].category))       

    return comply

其中category表示感兴趣的变量（例如Men或Employed或Part_time）。但我似乎无法以这种方式将category称为输入变量。怎么会这样做？

Answer 1

您可以使用df[category]

def cutoff(category, cut, direction):

    if direction == 0:
        comply = list(zip(df[df[category] < cut].Major, df[df[category] < cut][category]))

    if direction == 1:
        comply = list(zip(df[df[category] > cut].Major, df[df[category] > cut][category]))  

    return comply

Answer 2

您可以按属性或项目访问熊猫框架。必须在编码时知道属性，项必须是字符串，因此可以是变量。

df.Major

VS

df['Major']

我建议，将切断和转换分成列表：

def cutoff(df, category, cut, direction):
    mask = df[category] < cut if direction == 0 else df[category] > cut
    return df[mask]

def get_list(df, category):
    return list(zip(df.Major, df[category]))

get_list(cutoff(df, 'Employed', 1000, 1))

在Python中定义函数时，如何将标题称为参数？

2 个答案: