英文:
how to query dataframe using pandas with multiple row conditions?
问题
You can fix the issue by using parentheses to properly group your conditions in the DataFrame selection. Here's the corrected code:
df_status = df.loc[(df['build'] == 2979) & (df['status'] == 'passed')]
This will return all rows where 'build' is equal to 2979 and 'status' is equal to 'passed'.
英文:
I have the below data set, and I need to use pandas to query the status of build (2979). how many passed and how many failed
build status
0 2980 passed
1 2979 passed
2 2979 passed
3 2979 failed
4 2978 failed
.. ... ...
101 2968 failed
102 2968 passed
103 2968 passed
104 2968 passed
105 2968 failed
if I do:
df_status =df.loc[df['build'] == 2979]
it will return the rows of build == 2979. It is what I wanted
but when I do
df_status =df.loc[df['build'] == 2979]& df['status']=='passed']
it returns TypeError: Cannot perform 'rand_' with a dtyped [object] array and scalar of type [bool].
How do I fix the problem?
I have tried using :
df_status =df.loc[df['build'] == 2979 & df['status']=='passed']
and hope to return all row with build ==2979 and status == 'passed', but it return TypeError: Cannot perform 'rand_' with a dtyped [object] array and scalar of type [bool].
答案1
得分: 1
这是运算符优先级的问题。您忘了用括号括起您的表达式。
当您执行：
df.loc[df['build'] == 2979 & df['status'] == 'passed']
Python会尝试评估：2979 & df['status']，因为&运算符的优先级高于==运算符。
您应该这样做：
df.loc[(df['build'] == 2979) & (df['status'] == 'passed')]
以恢复两个操作的优先级。上面的代码等同于：
df.loc[df['build'].eq(2979) & df['status'].eq('passed')]
英文:
This is problem of operator precedence. You missed to enclose your expression with some parenthesis.
When you do:
df.loc[df['build'] == 2979 & df['status'] == 'passed']
Python try to evaluate: 2979 & df['status'] because & operator has an higher precedence than == operator.
What you should do is:
df.loc[(df['build'] == 2979) & (df['status'] == 'passed')]
To restore the priority of both operations. The code above is equivalent to:
df.loc[df['build'].eq(2979) & df['status'].eq('passed')]
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论