How to delete from a dataframe by condition python-CodePudding

I have a dataframe in pandas with columns QUESTIONS and ANSWERS

| QUESTION | ANSWER |
| -------- | ------ |
| www      | 123    |
| aaa      | 3546   |
| vvv      | 432    |
| ttt      | 455    |
| QUESTION | 534    |
| eee      | 4344   |
| yyy      | 5435   |

I need to delete a row = 'QUESTIONS' in column QUESTIONS I do it this 2 ways but it deletes the whole column

# 1 approach
test_df = test_df.drop("QUESTIONS", axis=1) 
 
# 2 approach
test_df = test_df.set_index("QUESTIONS")
test_df = test_df.drop("QUESTIONS", axis=0) # Delete all rows with label

What is my mistake that it deletes the whole column?

CodePudding user response：

Use the loc operator

test_df = test_df.loc[~(test_df.QUESTION=='QUESTION')]

CodePudding user response：

@Arnau's method is much better. If you need to use drop, you could use:

out = df.set_index('QUESTION').drop(['QUESTION']).reset_index()

or (which is just a convoluted way to write the method in @Arnau's answer):

out = df.drop(df.loc[df['QUESTION']=='QUESTION'].index)

Output:

  QUESTION  ANSWER
0      www     123
1      aaa    3546
2      vvv     432
3      ttt     455
4      eee    4344
5      yyy    5435