CODEWITHSUNDEEP
pythonpandaspython-3.5
Cy Bu
Cy Bu

Reputation: 1437

python pandas: find the element that is a string in a pandas DataFrame

Say i have a pandas DataFrame: 

df = pd.DataFrame({'a': [1,2,3,'e',4], 'b': [1,2,3,4,5]})

I would like to have the index of where the element of df is a string. How can i do that other than by checking element after element which slow and inefficient?

Upvotes: 1

Views: 1126

Answers (1)

MaxU - stand with Ukraine
MaxU - stand with Ukraine

Reputation: 210832

It's not exactly what you were asking. It rather returns you an index of elements that can't be converted to numeric values:

In [231]: df
Out[231]:
   a  b
0  1  1
1  2  2
2  3  3
3  e  4
4  4  5

In [232]: df.apply(pd.to_numeric, errors='coerce').isnull().any(1)
Out[232]:
0    False
1    False
2    False
3     True
4    False
dtype: bool

In [233]: df.loc[df.apply(pd.to_numeric, errors='coerce').isnull().any(1)]
Out[233]:
   a  b
3  e  4

Or more efficient variant from @Zero, which will check only string (object) columns:

In [237]: df.select_dtypes(['object']).apply(pd.to_numeric, errors='coerce').isnull().any(1)
Out[237]:
0    False
1    False
2    False
3     True
4    False
dtype: bool

In [238]: df[df.select_dtypes(['object']).apply(pd.to_numeric, errors='coerce').isnull().any(1)]
Out[238]:
   a  b
3  e  4

Upvotes: 2

Related Questions