将函数应用于pandas中数据框的每一列

I have this:

我有这个：

df = DataFrame(dict(person= ['andy', 'rubin', 'ciara', 'jack'], 
     item = ['a', 'b', 'a', 'c'], 
     group= ['c1', 'c2', 'c3', 'c1'], 
     age= [23, 24, 19, 49]))
df:

    age group item person
0   23  c1    a    andy
1   24  c2    b    rubin
2   19  c3    a    ciara
3   49  c1    c    jack

what I want to do, is to get the length of unique items in each column. Now I know I can do something like:

我想要做的是获得每列中唯一项目的长度。现在我知道我可以这样做：

len(df.person.unique())

for every column.

对于每一列。

Is there a way to do this in one go for all columns?

有没有办法一次性完成所有列？

I tried to do:

我试着做：

for column in df.columns:
    print(len(df.column.unique()))

but I know this is not right.

但我知道这不对。

How can I accomplish this?

我怎么能做到这一点？

3 个解决方案

#1

you want pd.Series.nunique

你想要pd.Series.nunique

df.apply(pd.Series.nunique)

age       4
group     3
item      3
person    4
dtype: int64

#2

You can the number of unique items in each column as:

您可以将每列中的唯一项目数量设置为：

for column in df.columns:
    print(len(df[column].unique()))

#3

You can use:

您可以使用：

for column in df:
    print(len(df[column].unique()))

4
3
3
4

Or:

要么：

for column in df:
    print(df[column].nunique())

4
3
3
4

#1