HOWTO · Pandas

Pandas DataFrame 选择列

本教程介绍了如何通过索引或使用 DataFrame.drop()和 DataFrame.filter()方法从 Pandas DataFrame 中选择列。

本页内容

本教程介绍了如何通过索引或使用 DataFrame.drop()DataFrame.filter() 方法从 Pandas DataFrame 中选择列。

我们将使用下面的 DataFrame df 来解释如何从 Pandas DataFrame 中选择列。

import pandas as pd

df = pd.DataFrame(
    {
        "A": [302, 504, 708, 103, 343, 565],
        "B": [100, 300, 400, 200, 400, 700],
        "C": [300, 400, 350, 100, 1000, 400],
        "D": [10, 15, 5, 0, 2, 7],
        "E": [4, 5, 6, 7, 8, 9],
    }
)

print(df)

输出:

     A    B     C   D  E
0  302  100   300  10  4
1  504  300   400  15  5
2  708  400   350   5  6
3  103  200   100   0  7
4  343  400  1000   2  8
5  565  700   400   7  9

使用索引操作从 Pandas DataFrame 中选择列

import pandas as pd

df = pd.DataFrame(
    {
        "A": [302, 504, 708, 103, 343, 565],
        "B": [100, 300, 400, 200, 400, 700],
        "C": [300, 400, 350, 100, 1000, 400],
        "D": [10, 15, 5, 0, 2, 7],
        "E": [4, 5, 6, 7, 8, 9],
    }
)

derived_df = df[["A", "C", "E"]]

print("The initial DataFrame is:")
print(df, "\n")

print("The DataFrame with A,C and E columns is:")
print(derived_df, "\n")

输出:

The initial DataFrame is:
     A    B     C   D  E
0  302  100   300  10  4
1  504  300   400  15  5
2  708  400   350   5  6
3  103  200   100   0  7
4  343  400  1000   2  8
5  565  700   400   7  9 

The DataFrame with A,C and E columns is:
     A     C  E
0  302   300  4
1  504   400  5
2  708   350  6
3  103   100  7
4  343  1000  8
5  565   400  9 

它从 DataFrame df 中选择列 ACE,并将这些列分配到 derived_df DataFrame 中。

使用 DataFrame.drop() 方法从 Pandas DataFrame 中选择列

import pandas as pd

df = pd.DataFrame(
    {
        "A": [302, 504, 708, 103, 343, 565],
        "B": [100, 300, 400, 200, 400, 700],
        "C": [300, 400, 350, 100, 1000, 400],
        "D": [10, 15, 5, 0, 2, 7],
        "E": [4, 5, 6, 7, 8, 9],
    }
)

derived_df = df.drop(["B", "D"], axis=1)

print("The initial DataFrame is:")
print(df, "\n")

print("The DataFrame with A,C and E columns is:")
print(derived_df, "\n")

输出:

The initial DataFrame is:
     A    B     C   D  E
0  302  100   300  10  4
1  504  300   400  15  5
2  708  400   350   5  6
3  103  200   100   0  7
4  343  400  1000   2  8
5  565  700   400   7  9 

The DataFrame with A,C and E columns is:
     A     C  E
0  302   300  4
1  504   400  5
2  708   350  6
3  103   100  7
4  343  1000  8
5  565   400  9 

它从 DataFrame df 中删除 BD 列,并将其余列分配给 derived_df。或者,它选择除 BD 以外的所有列,并将它们分配到 derived_df DataFrame 中。

使用 DataFrame.filter() 方法从 Pandas DataFrame 中选择列

import pandas as pd

df = pd.DataFrame(
    {
        "A": [302, 504, 708, 103, 343, 565],
        "B": [100, 300, 400, 200, 400, 700],
        "C": [300, 400, 350, 100, 1000, 400],
        "D": [10, 15, 5, 0, 2, 7],
        "E": [4, 5, 6, 7, 8, 9],
    }
)

derived_df = df.filter(["A", "C", "E"])

print("The initial DataFrame is:")
print(df, "\n")

print("The DataFrame with A,C and E columns is:")
print(derived_df, "\n")

输出:

The initial DataFrame is:
     A    B     C   D  E
0  302  100   300  10  4
1  504  300   400  15  5
2  708  400   350   5  6
3  103  200   100   0  7
4  343  400  1000   2  8
5  565  700   400   7  9 

The DataFrame with A,C and E columns is:
     A     C  E
0  302   300  4
1  504   400  5
2  708   350  6
3  103   100  7
4  343  1000  8
5  565   400  9

它从 DataFrame df 中提取或过滤 ACE 列,并将其分配给 DataFrame derived_df