python - 如何基于其他数据框在数据框中创建连接？

Question

我有 2 个数据框。一个包含学生批处理详细信息，另一个包含分数。我想加入 2 个数据框。

Dataframe1 包含

+-------+-------+-------+--+
|  s1   |  s2   |  s3   |  |
+-------+-------+-------+--+
| Stud1 | Stud2 | Stud3 |  |
| Stud2 | Stud4 | Stud1 |  |
| Stud1 | Stud3 | Stud4 |  |
+-------+-------+-------+--+

Dataframe2 包含

+-------+-------+----------+--+
| Name  | Point | Category |  |
+-------+-------+----------+--+
| Stud1 |    90 | Good     |  |
| Stud2 |    80 | Average  |  |
| Stud3 |    95 | Good     |  |
| Stud4 |    55 | Poor     |  |
+-------+-------+----------+

我正在尝试为每个学生在同一数据集中映射标记。

+-------+-------+-------+----+----+----+
| Stud1 | Stud2 | Stud3 | 90 | 80 | 95 |
| Stud2 | Stud4 | Stud1 | 80 | 55 | 90 |
| Stud1 | Stud3 | Stud4 | 90 | 95 | 55 |
+-------+-------+-------+----+----+----+

我试过下面的代码，但它正在一个一个地替换值。

s = df3['p1'].map(dfnamepoints.set_index('name')['points'])
df4 = df3.drop('p1', 1).assign(points = s)

Answer 1

如果 df3 中的所有值都存在于列 Name 中，解决方案同样有效:

s = dfnamepoints.set_index('Name')['Point']
df = df3.join(df3.replace(s).add_prefix('new_'))

或者:

df = df3.join(df3.apply(lambda x: x.map(s)).add_prefix('new_'))

或者:

df = df3.join(df3.applymap(s.get).add_prefix('new_'))

print (df)
      s1     s2     s3  new_s1  new_s2  new_s3
0  Stud1  Stud2  Stud3      90      80      95
1  Stud2  Stud4  Stud1      80      55      90
2  Stud1  Stud3  Stud4      90      95      55

如果不是，则输出不同 - 对于不存在的值 (Stud1) 获取 NaNs:

print (dfnamepoints)
    Name  Point Category
0  Stud2     80  Average
1  Stud3     95     Good
2  Stud4     55     Poor

df = df3.join(df3.applymap(s.get).add_prefix('new_'))
#or 
df = df3.join(df3.applymap(s.get).add_prefix('new_'))

print (df)
      s1     s2     s3  new_s1  new_s2  new_s3
0  Stud1  Stud2  Stud3     NaN      80    95.0
1  Stud2  Stud4  Stud1    80.0      55     NaN
2  Stud1  Stud3  Stud4     NaN      95    55.0

对于 replace 获取原始值:

df = df3.join(df3.replace(s).add_prefix('new_'))
print (df)
      s1     s2     s3 new_s1  new_s2 new_s3
0  Stud1  Stud2  Stud3  Stud1      80     95
1  Stud2  Stud4  Stud1     80      55  Stud1
2  Stud1  Stud3  Stud4  Stud1      95     55