如何使用逗号分隔的值列制作虚拟变量? [英] How to make dummy variables with comma separated valued columns?

问题描述

我正在为机器学习进行数据预处理，但是遇到了问题.
这是我想做的.

I am working on data preprocessing for machine learning and faced a problem.
Here is what I want to do.

表格图片:

表的类型为pandas数据框.

Table's type is pandas dataframe.

我的当前表是左表，我想将表转换为右表.

My current table is left one, and I want to transform my table to right one.

电影和演员的数目不是固定的.

The number of movies and actors are not fixed.

数据输入

 df=pd.DataFrame({'name':['A','B','C'],'actors':['a,b','b,d','c,m']})

预期输出:

       a  b  c  d  m

A      1  1  0  0  0
B      0  1  0  1  0
C      0  0  1  0  1

推荐答案

尝试一下? (顺便说一句，kaggle电影数据集，最好使用 LabelEncoder )

Try this ? (BTW , kaggle movie dataset, better using LabelEncoder)

PS:我没有添加列name，您可以简单地执行out['name']=df.name

PS: I did not add the column name, you can simply do out['name']=df.name

选项1 pd.crosstab

df.actors=df.actors.str.split(',')
df1=df.set_index('name').actors.apply(pd.Series).stack()
pd.crosstab(df1.index.get_level_values(0),df1).rename_axis(None).rename_axis(None,1)


Out[246]: 
   a  b  c  d  m
A  1  1  0  0  0
B  0  1  0  1  0
C  0  0  1  0  1

---

选项2 get_dummies

---

Option 2 get_dummies

pd.get_dummies(df.actors.str.split(',').apply(pd.Series).stack()).sum(level=0)

Out[230]: 
   a  b  c  d  m
0  1  1  0  0  0
1  0  1  0  1  0
2  0  0  1  0  1

---

选项3 MultiLabelBinarizer

---

Option 3 MultiLabelBinarizer

from sklearn.preprocessing import MultiLabelBinarizer
mlb = MultiLabelBinarizer()
    pd.DataFrame(mlb.fit_transform(df.actors.str.split(',')),columns=mlb.classes_,index=df.name).reset_index()
Out[238]: 
  name  a  b  c  d  m
0    A  1  1  0  0  0
1    B  0  1  0  1  0
2    C  0  0  1  0  1

---

数据输入

---

Data Input

df=pd.DataFrame({'name':['A','B','C'],'actors':['a,b','b,d','c,m']})

这篇关于如何使用逗号分隔的值列制作虚拟变量?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！