CODEWITHSUNDEEP
pythonreplacepandasdataframe
Ashleigh Clayton
Ashleigh Clayton

Reputation: 1445

Replace row in DataFrame dependant on a value

I would like to replace an entire row in a data, subject to one column in that row satisfying the condition. i.e. where red is the dataframe

['1' 0.0 ' P']
['2' 0.0 ' S']
['3' 64  ' M']
['4' 70  ' M']



red=red.replace(to_replace=' M', value=0)

returns the result:

['1' 0.0 ' P']
['2' 0.0 ' S']
['3' 64  0]
['4' 70  0]

but I would like it to return:

['1' 0.0 ' P']
['2' 0.0 ' S']
['3' 0  0]
['4' 0  0]

Upvotes: 1

Views: 1763

Answers (1)

Zeugma
Zeugma

Reputation: 32095

Use loc to filter out the part of the DataFrame you want to zero, and then assign the value to it. Below, it selects all lines where c column value is 'M' and it takes all columns from b to c, and set the value of this selection to 0:

df = pd.DataFrame([['1', 0.0, 'P'],
    ...: ['2', 0.0, 'S'],
    ...: ['3', 64,  'M'],
    ...: ['4', 70,  'M'],], columns=['a', 'b', 'c'])

df.loc[df['c']=='M','b':'c'] = 0

df
Out[54]: 
   a  b  c
0  1  0  P
1  2  0  S
2  3  0  0
3  4  0  0

Upvotes: 6

Related Questions