有没有一种方法可以对熊猫的数据帧中的一组重复项目进行编号？

问题描述

我有一个包含区域，客户和某些交货的数据框。 此列用作购买类型，并且首次购买和最后一次购买被标记为“第一笔”和“最后一次”，但是介于两者之间的任何交付均被标记为“交付” 有没有一种方法可以转换投放并获得“ delivery1”，“ delivery2”等标签？

import pandas as pd  
data = [['NY','A','FirsT',10],['NY','DELIVERY',20],30],'LAST',25],'B',15],['FL',12],20]
       ] 
  
# Create the pandas DataFrame 
df = pd.DataFrame(data,columns = ['Region','Client','purchaseType','price']) 
  
# print dataframe. 
df 

所需的输出：

data2 = [['NY','DELIVERY1','DELIVERY2','DELIVERY3',20]
       ] 

df2 = pd.DataFrame(data2,'price']) 
print(df2)

谢谢！

解决方法

我们可以尝试使用GroupBy.cumcount和Series.str.cat

blocks = df['purchaseType'].eq('FIRST').cumsum()
fill_values = df['purchaseType'].str.cat(df.groupby(blocks)
                                           .cumcount().astype(str),sep='')
df.loc[df['purchaseType'].eq('DELIVERY'),'purchaseType'] = fill_values
print(df)

#    Region Client purchaseType  price
# 0      NY      A        FIRST     10
# 1      NY      A    DELIVERY1     20
# 2      NY      A    DELIVERY2     30
# 3      NY      A         LAST     25
# 4      NY      B        FIRST     15
# 5      NY      B    DELIVERY1     10
# 6      NY      B         LAST     20
# 7      FL      A        FIRST     15
# 8      FL      A    DELIVERY1     10
# 9      NY      A    DELIVERY2     12
# 10     NY      A    DELIVERY3     25
# 11     NY      A         LAST     20

,

您可以使用np.where来决定在何处添加数字后缀：

df['purchaseType'] = df.groupby((df['purchaseType']=='FIRST').cumsum())['purchaseType'].transform(
    lambda x: np.where(x=='DELIVERY',x+np.arange(len(x)).astype(str),x)
)

print(df)

打印：

   Region Client purchaseType  price
0      NY      A        FIRST     10
1      NY      A    DELIVERY1     20
2      NY      A    DELIVERY2     30
3      NY      A         LAST     25
4      NY      B        FIRST     15
5      NY      B    DELIVERY1     10
6      NY      B         LAST     20
7      FL      A        FIRST     15
8      FL      A    DELIVERY1     10
9      NY      A    DELIVERY2     12
10     NY      A    DELIVERY3     25
11     NY      A         LAST     20