Python: Eliminating rows in Pandas DataFrame based on boolean condition

Question

Suppose I have a DataFrame in Pandas like

   c1   c2
0  'ab'  1
1  'ac'  0
2  'bd'  0
3  'fa'  1
4  'de'  0

and I want it to show all rows such that c1 doesn't contain 'a'. My desired output would be:

   c1   c2
2  'bd'  0
4  'de'  0

My first attempt was to use df.loc, like this:

df.loc['a' not in df['c1']]

For searching specific values, df.loc works fine, but for searching based on a False condition ('a' not in df['c1']) it doesn't.

I know I can do the reverse thing. I mean, i can return all rows which contain 'a' in column 'c1', through this code:

df.loc[df['c1'].str.contains('a')]

But I just can't figure out an elegant/concise way to do the other way around. How can I do that?

TomAugspurger · Accepted Answer · 2015-06-03 03:02:10Z

3

Use ~ to flip your Series of booleans:

df.loc[~df['c1'].str.contains('a')]

answered Jun 3, 2015 at 3:02

TomAugspurger

29k8 gold badges89 silver badges71 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Works also without the ".loc" to condense it even more: df[~df['c1'].str.contains('a')]