Question

我有一个数据框：

id    time
Uk6   year
36h   year
Uk6   two-year
rf5   month
gg7   year
rf5   half-year

我需要删除基于“ id”列的重复项，并将重复行的时间值替换为“ unknown”。结果应该是：

id      time
Uk6    unknown
36h    year
rf5    unknown
gg7    year

我尝试了对先前问题（like_this）的建议答案，但它们不起作用。

Answer 1

尝试以下

array = []
for l in self.pairs
    string = "%d - %d" % (self.ID, l)
array.append(string)
print ", ".join(array)


1-2, 1-3    
2-1, 2-4    
3-1, 3-4    
4-2, 4-3

输出

# create the dataframe
df = pd.DataFrame(data={'id': ['Uk6', '36h', 'Uk6', 'rf5', 'gg7', 'rf5'],
                        'time': ['year', 'year', 'two-year', 'month', 'year', 'half-year']})

# get duplicated id's
dups_id = df[df.duplicated(subset='id')]['id']

# remove rows from dataframe with id that has duplicated rows
df = df.drop_duplicates(subset='id')

# replace values of 'time' for those rows with duplicated id's with 'unknown'
df.loc[:,'time'] = df['time'].where(~df['id'].isin(dups_id), other='unknown')

Answer 2

您可以先获取重复项的索引，然后将相应的time值替换为unknown，最后删除重复项：

import pandas as pd

df = pd.DataFrame({'id': ["Uk6", "36h", "Uk6", "rf5", "gg7", "rf5"],
                   'time': ["year", "year", "two-year", "month", "year", "half-year"]})

mask = df.duplicated(subset= 'id', keep=False)
df['time'][mask] = "unknown"
df = df.drop_duplicates('id')

Answer 3

使用loc替换未知和删除重复的测试

df.loc[df.id.duplicated(keep = False), 'time'] = 'unknown'
df = df.drop_duplicates()

    id  time
0   Uk6 unknown
1   36h year
3   rf5 unknown
4   gg7 year

删除重复项后替换列值

3 个答案: