使用Apply将Series存储到Pandas DataFrame单元中

Question

I have a function that returns a Series. I apply the function to my DataFrame using apply(), and I would like to store the resulting Series into a new column of the dataframe. eg,

   | 1 |                  2
---+---+------------------------------------
 0 | 1 | <class 'pandas.core.series.Series'>
 1 | 2 | <class 'pandas.core.series.Series'>
 2 | 3 | <class 'pandas.core.series.Series'>

However, it seems that pandas wants to expand the series and store it into multiple columns, so I get a ValueError: Wrong number of items passed 3, placement implies 1.

import pandas as pd

df = pd.DataFrame({1: [1, 2, 3]})


def f(x):
    return pd.Series(['a', 'b', 'c'])

df[2] = df.apply(f, axis=1)

How should deal with this?

Answer 1

我不建议您将series保存到数据框单元格中，但这是转换为对象的一种方法

df['2']=[pd.Series(['a', 'b', 'c'])]*len(df)