Pandas分组groupby结合agg-transform

groupby结合agg和transform使用

本文介绍的是分组groupby分组之后如何使用agg和transform

模拟数据

import pandas as pd
import numpy as np

employees = ["小明","小周","小孙","小王","小张"]   # 5位员工
time = ["上半年", "下半年"]


df=pd.DataFrame({
    "employees":np.random.choice(employees,10),  # 在员工中重复选择10次
    # 另一种写法
    #"employees":[employees[x] for x in np.random.randint(0,len(employees),10)],
    "time":np.random.choice(time,10),
    "salary":np.random.randint(800,1000,10),  # 800-1000之间的薪资选择10个数值
    "score":np.random.randint(6,12,10)  # 6-11的分数选择10个
})

df

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

groupby+单个字段+单个聚合

求解每个人的总薪资金额：

total_salary = df.groupby("employees")["salary"].sum().reset_index()
total_salary

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

使用agg也能够实现上面的效果：

df.groupby("employees").agg({"salary":"sum"}).reset_index()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

df.groupby("employees").agg({"salary":np.sum}).reset_index()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

groupby+单个字段+多个聚合

求解每个人的总薪资金额和薪资的平均数：

方法1：使用groupby+merge

mean_salary = df.groupby("employees")["salary"].mean().reset_index()
mean_salary

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

然后将上面的两个结果进行组合；在合并之前为了字段的名字更加的直观，我们重命名下：

total_salary.rename(columns={"employees":"total_salary"})
mean_salary.columns = ["employees","mean_salary"]

total_mean = total_salary.merge(mean_salary)
total_mean

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

方法2：使用groupby+agg

total_mean = df.groupby("employees")\
            .agg(total_salary=("salary", "sum"),
                 mean_salary=("salary", "mean"))\
            .reset_index()
total_mean

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

groupby+多个字段+单个聚合

针对多个字段的同时聚合：

df.groupby(["employees","time"])["salary"].sum().reset_index()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

# 使用agg来实现

df.groupby(["employees","time"]).agg({"salary":"sum"}).reset_index()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>

groupby+多个字段+多个聚合

使用的方法是：

agg(’新列名‘=(’原列名‘, ’统计函数/方法‘))

df.groupby(["employees","time"])\
            .agg(total_salary=("salary", "sum"),
                mean_salary=("salary", "mean"),
                total_score=("score", "sum")
                )\
            .reset_index()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } <pre><code>.dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </code></pre>