跳转至

熊猫数据框中两列的差异

原文:https://www.geesforgeks.org/两栏之差-熊猫-数据框/

Python 熊猫数据框中两列的差异是通过以下方法实现的:

方法#1 : 使用“-”运算符。

import pandas as pd

# Create a DataFrame
df1 = { 'Name':['George','Andrea','micheal',
                'maggie','Ravi','Xien','Jalpa'],
        'score1':[62,47,55,74,32,77,86],
        'score2':[45,78,44,89,66,49,72]}

df1 = pd.DataFrame(df1,columns= ['Name','score1','score2'])

print("Given Dataframe :\n", df1)

# getting Difference
df1['Score_diff'] = df1['score1'] - df1['score2']
print("\nDifference of score1 and score2 :\n", df1)

Output:

Given Dataframe :
       Name  score1  score2
0   George      62      45
1   Andrea      47      78
2  micheal      55      44
3   maggie      74      89
4     Ravi      32      66
5     Xien      77      49
6    Jalpa      86      72

Difference of score1 and score2 :
       Name  score1  score2  Score_diff
0   George      62      45          17
1   Andrea      47      78         -31
2  micheal      55      44          11
3   maggie      74      89         -15
4     Ravi      32      66         -34
5     Xien      77      49          28
6    Jalpa      86      72          14

方法#2 : 使用子()方法的数据框。

import pandas as pd

# Create a DataFrame
df1 = { 'Name':['George','Andrea','micheal',
                'maggie','Ravi','Xien','Jalpa'],
        'score1':[62,47,55,74,32,77,86],
        'score2':[45,78,44,89,66,49,72]}

df1 = pd.DataFrame(df1,columns= ['Name','score1','score2'])

print("Given Dataframe :\n", df1)

df1['Score_diff'] = df1['score1'].sub(df1['score2'], axis = 0)
print("\nDifference of score1 and score2 :\n", df1)

Output:

Given Dataframe :
       Name  score1  score2
0   George      62      45
1   Andrea      47      78
2  micheal      55      44
3   maggie      74      89
4     Ravi      32      66
5     Xien      77      49
6    Jalpa      86      72

Difference of score1 and score2 :
       Name  score1  score2  Score_diff
0   George      62      45          17
1   Andrea      47      78         -31
2  micheal      55      44          11
3   maggie      74      89         -15
4     Ravi      32      66         -34
5     Xien      77      49          28
6    Jalpa      86      72          14



回到顶部