CODEWITHSUNDEEP
pythonpython-3.xpandasdataframenan
NWWPA
NWWPA

Reputation: 59

Pandas replacing all column values with nan

I have a pandas dataframe that contains a column populated by 3 object values: "true","false", and "---"

I want to replace the "true" values with 0, and the "false" values with 1, so I wrote the following

df['True/False'] = df['True/False'].str.replace('False' '1')
df['True/False'] = df['True/False'].str.replace('True', '0')

So, I would expect input ['False','True','False','False','False','---'] to return [1,0,1,1,1,'---'] when I print(df['True/False']).

But instead I get:

["NaN","NaN","NaN","NaN","NaN","NaN"]

A little help?

Upvotes: 1

Views: 198

Answers (1)

Mayank Porwal
Mayank Porwal

Reputation: 34086

Instead of df.replace, you should use Series.map as it is much faster:

Create a dict of values you want to replace:

In [3590]: d = {'True': 0, 'False': 1}

Consider df:

In [3592]: df = pd.DataFrame(['False','True','False','False','False','---'], columns=['A'])

In [3593]: df
Out[3593]: 
       A
0  False
1   True
2  False
3  False
4  False
5    ---

In [3597]: df.A = df.A.map(d).fillna(df.A)

In [3598]: df
Out[3598]: 
     A
0    1
1    0
2    1
3    1
4    1
5  ---

Upvotes: 3

Related Questions