CODEWITHSUNDEEP
pythonpandas
silvercoder
silvercoder

Reputation: 149

Identify integers in column and replace with string

I'm new to python and pandas. I have a dataframe that looks as follows:

             ID                    NAME
0      0000001                    Apple
1      0000002                       35
2      0000003                    Grape
3      0000004                       22
4      0000005                   Banana
5      0000006                       12

My goal is to replace the fields with numbers in the NAME column with 'Unknown'

So far, I have tried the following:

out['NAME'] = out.apply(lambda x: x['NAME'].replace(x['NAME'], 'Unknown'))

But it wouldn't seem to replace it and gives me a KeyError: ('NAME', 'occurred at index ID')

Ultimately I am expecting an output as follows:

             ID                    NAME
0      0000001                    Apple
1      0000002                   Unkown
2      0000003                    Grape
3      0000004                   Unkown
4      0000005                   Banana
5      0000006                   Unkown

Upvotes: 1

Views: 185

Answers (4)

AbelianGrapes
AbelianGrapes

Reputation: 31

You can also just create a function in python

def replaceName(Name):
 if Name.isnumeric():
  return "Unknown"
 else:
  return Name

and use the map function:

df["Name"] = df["Name"].map(lambda x: replaceName(x))

Upvotes: 0

Pablo C
Pablo C

Reputation: 4761

You can use pandas.to_numeric to determinate which rows can be converted to numeric:

>>> df.loc[~pd.to_numeric(df["NAME"], errors = "coerce").isnull(), "NAME"] = "unknown"
>>> df
   ID     NAME
0   1    Apple
1   2  unknown
2   3    Grape
3   4  unknown
4   5   Banana
5   6  unknown

With errors = "coerce" the invalid parsing will be set to NaN.

Upvotes: 0

wwnde
wwnde

Reputation: 26676

out['NAME']=np.where(out.NAME.str.contains('\d'),'unknown',out.NAME)

Upvotes: 0

Sayandip Dutta
Sayandip Dutta

Reputation: 15872

Use pandas.Series.str.isnumeric:

>>> out.loc[out.NAME.str.isnumeric(), 'NAME'] = 'Unknown'
>>> out

        ID     NAME
0  0000001    Apple
1  0000002  Unknown
2  0000003    Grape
3  0000004  Unknown
4  0000005   Banana
5  0000006  Unknown

Upvotes: 1

Related Questions