Question

使用Pandas，获取以下数据集

author1,category1,10.00
author1,category2,15.00
author1,category3,12.00
author2,category1,5.00
author2,category2,6.00
author2,category3,4.00
author2,category4,9.00
author3,category1,7.00
author3,category2,4.00
author3,category3,7.00

我想为每位作者获得最高价值

author1,category2,15.00
author2,category4,9.00
author3,category1,7.00
author3,category3,7.00

（道歉，我是一只大熊猫。）

Answer 1

import pandas as pd

df = pd.read_csv("in.csv", names=("Author","Cat","Val"))

print(df.groupby(['Author'])['Val'].max())

获取df：

inds = df.groupby(['Author'])['Val'].transform(max) == df['Val']
df = df[inds]
df.reset_index(drop=True, inplace=True)
print(df)
    Author        Cat  Val
0  author1  category2   15
1  author2  category4    9
2  author3  category1    7
3  author3  category3    7

Answer 2

由于您还要检索category列，因此.agg列上的标准val无法满足您的需求。（另外，由于author3中有两个值为7，因此@Padraic Cunningham使用.max()的方法只返回一个实例而不是两个实例。您可以定义一个自定义的apply函数来完成任务。

import pandas as pd

# your data, assume columns names are: author, cat, val
# ===============================
print(df)


    author        cat  val
0  author1  category1   10
1  author1  category2   15
2  author1  category3   12
3  author2  category1    5
4  author2  category2    6
5  author2  category3    4
6  author2  category4    9
7  author3  category1    7
8  author3  category2    4
9  author3  category3    7

# processing
# ====================================
def func(group):
    return group.loc[group['val'] == group['val'].max()]

df.groupby('author', as_index=False).apply(func).reset_index(drop=True)


    author        cat  val
0  author1  category2   15
1  author2  category4    9
2  author3  category1    7
3  author3  category3    7

使用Python中的Pandas，为每个组选择最高值行

2 个答案: