CODEWITHSUNDEEP
pythonpandasdataframe
rosefun
rosefun

Reputation: 1857

How to check if a dataframe contains a certain element?

I have a DataFrame named df , and I want to know the df whether contains the element a in each row.

import pandas as pd
import numpy as np 
df=pd.DataFrame({'id':[1,2,3],'item1':['a','c','a'],
'item2':['b','d','e'],'item3':['c','e',np.nan]})

Input:

   id item1 item2 item3
0   1     a     b     c
1   2     c     d     e
2   3     a     e   NaN

In the new column contain_a , 1 represents the column item1 or item2 or item3 contains the element a .

Expected:

   id item1 item2 item3 contains_a
0   1     a     b     c         1
1   2     c     d     e         0
2   3     a     e   NaN         1

Upvotes: 0

Views: 147

Answers (2)

Alexander
Alexander

Reputation: 109546

Check each column barring id in column 1 (df.iloc[:, 1:]) using the string accessor to see if it contains the letter a, and then use any along the rows (axis=1). Convert the boolean result to an integer (1 or 0).

>>> df.assign(contains_a=df.iloc[:, 1:].apply(lambda s: s.str.contains('a')).any(axis=1).astype(int))
   id item1 item2 item3  contains_a
0   1     a     b     c           1
1   2     c     d     e           0
2   3     a     e   NaN           1

Too make this more general for multiple targets:

targets = ['aa', 'a', 'b', 'c']
d = {'contains_{}'.format(target): 
         df.iloc[:, 1:].apply(lambda s: s.str.contains(target)).any(axis=1).astype(int)
     for target in targets}

>>> df.assign(**d)
   id item1 item2 item3  contains_a  contains_aa  contains_b  contains_c
0   1     a     b     c           1            0           1           1
1   2     c     d     e           0            0           0           1
2   3     a     e   NaN           1            0           0           0

Upvotes: 1

Zero
Zero

Reputation: 76917

Use

In [578]: df['contains_a'] = df.filter(like='item').eq('a').any(1).astype(int)

In [579]: df
Out[579]:
   id item1 item2 item3  contains_a
0   1     a     b     c           1
1   2     c     d     e           0
2   3     a     e   NaN           1

Upvotes: 2

Related Questions