Is there a better way to select rows from a pandas DataFrame based on multiple conditions?

Question

I'm trying to select rows from a pandas DataFrame based on multiple conditions. The code looks like this:

row = videos_train_df[
             (videos_train_df['pid1']==pid1)
            &(videos_train_df['pid2']==pid2)
            &(videos_train_df['vid'] ==vid)]

Is there any better way (in terms of code readability) to do the same thing?

https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.query.html ? — MaxNoe, Dec 23 '21 at 21:54
Do you find that difficult to read? I don't, especially the way you've spaced it. I think the meaning is quite clear. — Tim Roberts, Dec 23 '21 at 21:54
@Tim Roberts this particular example isn't too difficult to read, but `df.query` is a nice way to avoid reusing the `videos_train_df` variable name in the condition. Plus there could be 100 conditions instead of 3 conditions and then passing a string to df.query is easier to work with — Derek O, Dec 23 '21 at 21:57

score 3 · Accepted Answer · answered Dec 23 '21 at 21:55

3

You can use query

row = videos_train_df.query(
    f"pid1 == {pid1} and pid2 == {pid2} and vid == {vid}"
)

See also this question.

answered Dec 23 '21 at 21:55

Andrea Di Iura

467
5
11

That doesn't need quoting? – Tim Roberts Dec 23 '21 at 21:59
Thanks, this is what I was looking for – Abdulwahab Almestekawy Dec 23 '21 at 22:06

score 0 · Answer 2 · answered Dec 23 '21 at 22:05

0

I will do all after eq

new = df[df[['pid1','pid2','vid']].eq([pid1,pid2,vid]).all(1)]

answered Dec 23 '21 at 22:05

BENY

317,841
20
164
234

Is there a better way to select rows from a pandas DataFrame based on multiple conditions?

2 Answers2