Remove row from pandas dataframe if it has 2 strings

Question

Im looking for a way to remove a row from a pandas data frame if it contains both of the 2 strings. I can do it if it has one, but havent been able to get both to work. Below is the code i used to remove records based on 1 string, looking to change that to include another keyword

code:

Vikings_dataframe_cleaned2=Vikings_dataframe_cleaned[Vikings_dataframe_cleaned.TweetText.str.contains("RT") == False]

Strings to be excluded: I want to check to text to make sure that it doesnt contain both @teddyb_h20 and @casekeenum7.

Example:

['@teddyb_h2o test test','@casekeenum7 and @teddyb_h2o are test','@casekeenum7 is the best right now']

The code should then produce a dataframe that looks like this:

['@teddyb_h2o test test','@casekeenum7 is the best right now']

Its a dupe, use Vikings_dataframe_cleaned[~Vikings_dataframe_cleaned.TweetText.str.contains("RT|'Your_other_string") ] — Vaishali, Dec 09 '17 at 00:45
Possible duplicate of [pandas + dataframe - select by partial string](https://stackoverflow.com/q/11350770/1278112) — Shihe Zhang, Dec 09 '17 at 01:03
I just saw the message trail, can you post a sample of your dataframe and what strings do you want to exclude? — Vaishali, Dec 09 '17 at 01:44
I just posted, hopefully that will help. Let me know if you need more — J. McCraiton, Dec 09 '17 at 01:52
The dataframe only consists of the 1 column, then i add a blank column that I use in R. But i have updated the question — J. McCraiton, Dec 09 '17 at 02:00
For my example wouldnt I use & to make sure that it contains both of the @'s — J. McCraiton, Dec 09 '17 at 02:04

Vaishali · Accepted Answer · 2017-12-09T02:14:49.027

2

Sample df

df = pd.DataFrame({'col': ['@teddyb_h2o test test','@casekeenum7 and @teddyb_h2o are test','@casekeenum7 is the best right now','test test']})

    col
0   @teddyb_h2o test test
1   @casekeenum7 and @teddyb_h2o are test
2   @casekeenum7 is the best right now
3   test test

Solution:

df[~(df.col.str.contains('@teddyb_h2o') & df.col.str.contains('@casekeenum7'))]

    col
0   @teddyb_h2o test test
2   @casekeenum7 is the best right now
3   test test

@Wen's suggestion, more elegant

df[~df['col'].str.contains(r'^(?=.*@teddyb_h2o)(?=.*@casek‌eenum7)')]

edited Dec 09 '17 at 02:14

answered Dec 09 '17 at 02:11

Vaishali

32,439
4
39
71

I got downvote for this solution ...good luck ...upvote for yours :-) – BENY Dec 09 '17 at 02:12
I saw, trying to figure out. Please undelete, i will delete mine – Vaishali Dec 09 '17 at 02:12
1

No need , also if you do not mind you can add ..`df[~df['col'].str.contains(r'^(?=.*@teddyb_h2o)(?=.*@casekeenum7)')]` :-) – BENY Dec 09 '17 at 02:13
Perfect! This worked. Thanks for all your help everyone! :) – J. McCraiton Dec 09 '17 at 02:14
Thats even better! You should answer:) – Vaishali Dec 09 '17 at 02:14
3

Let us keep this site straightforward to the right answers :-) – BENY Dec 09 '17 at 02:14

Remove row from pandas dataframe if it has 2 strings

1 Answers1