pandas dataframe dataframe.merge() 函数-ag捕鱼王app官网

python pandas dataframe.merge() 函数合并 dataframe 或命名的 series 对象。

pandas.dataframe.merge() 语法

dataframe.merge(
    right,
    how="inner",
    on=none,
    left_on=none,
    right_on=none,
    left_index=false,
    right_index=false,
    sort=false,
    suffixes="_x",
    "_y",
    copy=true,
    indicator=false,
    validate=none,
)

参数

返回值

它返回一个合并给定对象的 dataframe。

示例代码：dataframe.merge() 函数合并两个 dataframe

import pandas as pd
df1 = pd.dataframe(
    {"name": ["suraj", "zeppy", "alish", "sarah"], "working hours": [1, 2, 3, 5]}
)
df2 = pd.dataframe({"name": ["suraj", "zack", "alish", "raphel"], "pay": [5, 6, 7, 8]})
print("1st dataframe:")
print(df1)
print("2nd dataframe:")
print(df2)
merged_df = df1.merge(df2)
print("merged dataframe:")
print(merged_df)

输出：

1st dataframe:
    name  working hours
0  suraj              1
1  zeppy              2
2  alish              3
3  sarah              5
2nd dataframe:
     name  pay
0   suraj    5
1    zack    6
2   alish    7
3  raphel    8
merged dataframe:
    name  working hours  pay
0  suraj              1    5
1  alish              3    7

它使用 sql 的内连接技术将 df1 和 df2 合并为一个 dataframe。

对于 inner-join 方法，我们必须确保两个 dataframe 至少有一列是共同的。

在这里，merge() 函数将把具有相同值的公共列的行连接到两个 dataframe。

示例代码：在 merge 方法中设置 how 参数，使用各种技术合并 dataframe

import pandas as pd
df1 = pd.dataframe(
    {"name": ["suraj", "zeppy", "alish", "sarah"], "working hours": [1, 2, 3, 5]}
)
df2 = pd.dataframe({"name": ["suraj", "zack", "alish", "raphel"], "pay": [5, 6, 7, 8]})
print("1st dataframe:")
print(df1)
print("2nd dataframe:")
print(df2)
merged_df = df1.merge(df2, how="right")
print("merged dataframe:")
print(merged_df)

输出：

1st dataframe:
    name  working hours
0  suraj              1
1  zeppy              2
2  alish              3
3  sarah              5
2nd dataframe:
     name  pay
0   suraj    5
1    zack    6
2   alish    7
3  raphel    8
merged dataframe:
     name  working hours  pay
0   suraj            1.0    5
1   alish            3.0    7
2    zack            nan    6
3  raphel            nan   8 

它使用 sql 的 right-join 技术将 df1 和 df2 合并为一个 dataframe。

在这里，merge() 函数从右边的 dataframe 返回所有的行。然而，只存在于左侧 dataframe 中的行将得到 nan 值。

同样，我们也可以使用 how 参数的 left 和 outer 值。

示例代码：在 pandas 中使用 dataframe.merge() 函数只合并特定的列

import pandas as pd
df1 = pd.dataframe(
    {
        "name": ["suraj", "zeppy", "alish", "sarah"],
        "working hours": [1, 2, 3, 5],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
df2 = pd.dataframe(
    {
        "name": ["suraj", "zack", "alish", "raphel"],
        "pay": [5, 6, 7, 8],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
print("1st dataframe:")
print(df1)
print("2nd dataframe:")
print(df2)
merged_df = df1.merge(df2, on="name")
print("merged dataframe:")
print(merged_df)

输出：

1st dataframe:
    name  working hours    position
0  suraj              1    salesman
1  zeppy              2         ceo
2  alish              3     manager
3  sarah              5  sales head
2nd dataframe:
     name  pay    position
0   suraj    5    salesman
1    zack    6         ceo
2   alish    7     manager
3  raphel    8  sales head
merged dataframe:
    name  working hours position_x  pay position_y
0  suraj              1   salesman    5   salesman
1  alish              3    manager    7    manager

它只合并 df1 和 df2 的 name 列。由于默认的连接方法是内连接，因此只有两个 dataframe 的共同行才会被连接。position 列是两个 dataframe 共同的，因此有两个位置列，即 position_x 和 position_y。

默认情况下，_x 和 _y 后缀被附加到重叠列的名称中。我们可以使用 suffixes 参数指定后缀。

df1 = pd.dataframe(
    {
        "name": ["suraj", "zeppy", "alish", "sarah"],
        "working hours": [1, 2, 3, 5],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
df2 = pd.dataframe(
    {
        "name": ["suraj", "zack", "alish", "raphel"],
        "pay": [5, 6, 7, 8],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
print("1st dataframe:")
print(df1)
print("2nd dataframe:")
print(df2)
merged_df = df1.merge(df2, on="name", suffixes=("_left", "_right"))
print("merged dataframe:")
print(merged_df)

输出：

1st dataframe:
    name  working hours    position
0  suraj              1    salesman
1  zeppy              2         ceo
2  alish              3     manager
3  sarah              5  sales head
2nd dataframe:
     name  pay    position
0   suraj    5    salesman
1    zack    6         ceo
2   alish    7     manager
3  raphel    8  sales head
merged dataframe:
    name  working hours position_left  pay position_right
0  suraj              1      salesman    5       salesman
1  alish              3       manager    7        manager

示例代码:使用索引作为连接键来合并 dataframe

import pandas as pd
df1 = pd.dataframe(
    {
        "name": ["suraj", "zeppy", "alish", "sarah"],
        "working hours": [1, 2, 3, 5],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
df2 = pd.dataframe(
    {
        "name": ["suraj", "zack", "alish", "raphel"],
        "pay": [5, 6, 7, 8],
        "position": ["salesman", "ceo", "manager", "sales head"],
    }
)
print("1st dataframe:")
print(df1)
print("2nd dataframe:")
print(df2)
merged_df = df1.merge(
    df2, left_index=true, right_index=true, suffixes=("_left", "_right")
)
print("merged dataframe:")
print(merged_df)

输出：

1st dataframe:
    name  working hours    position
0  suraj              1    salesman
1  zeppy              2         ceo
2  alish              3     manager
3  sarah              5  sales head
2nd dataframe:
     name  pay    position
0   suraj    5    salesman
1    zack    6         ceo
2   alish    7     manager
3  raphel    8  sales head
merged dataframe:
  name_left  working hours position_left name_right  pay position_right
0     suraj              1      salesman      suraj    5       salesman
1     zeppy              2           ceo       zack    6            ceo
2     alish              3       manager      alish    7        manager
3     sarah              5    sales head     raphel    8     sales head

它合并两个 dataframe 的相应行，不考虑列的相似性。如果两个 dataframe 上出现相同的列名，则在合并后将后缀附加到列名上，使之成为不同的列。