CODEWITHSUNDEEP
pythonpandasdataframe
alim1990
alim1990

Reputation: 4972

Python how can I remove the floating point from numbers included within a field having string values?

I have a dataframe containing fields having numbers with string and empty values like the in the following:

import pandas as pd
import numpy as np
df = pd.DataFrame()
new_row = [{'Col1': 1.0, 'Col2': 'c', 'Col3': 1.02}, {'Col1': 1.0, 'Col2': 'e', 'Col3': ''}]
df = df.append(new_row, ignore_index=True)

The dataframe:

Col1    Col2    Col3

1.0     c       2.0

1.0     e

I need to remove .0 from fields having strings as well, so the dataframe should look like this:

Col1    Col2    Col3

1.0     c       2

1.0     e

It is not only for col3 because I have a huge real life dataframe. I tried the following:

for col in df.columns:
    print(col)
    df[col] = df[col].apply(lambda x: f'{x: .0f}')

And it returned an error:

ValueError: Unknown format code 'f' for object of type 'str'

So how can I remove the floating point from numbers included within a field having string values?

Upvotes: 0

Views: 470

Answers (4)

jezrael
jezrael

Reputation: 862641

If need replace .0 only for columns with mixed values strings and numbers with .0 use:

#data was changed for match sample data with 2.0 in Col3
df = pd.DataFrame()
new_row = [{'Col1': 1.0, 'Col2': 'c', 'Col3': 2.0}, {'Col1': 1.0, 'Col2': 'e', 'Col3': ''}]
df = df.append(new_row, ignore_index=True)

c = df.select_dtypes(object).columns

df[c] = df[c].astype(str).replace('\.0$','', regex=True)
print (df)
   Col1 Col2 Col3
0   1.0    c    2
1   1.0    e

If need processing all values:

def f(x):
    try:
        return f'{x: .0f}'
    except:
        return x

df = df.applymap(f)
print (df)
  Col1 Col2 Col3
0    1    c    2
1    1    e

Solution with test floats like integers (end by .0):

df = pd.DataFrame()
new_row = [{'Col1': 1.15, 'Col2': 'c', 'Col3': 2.0}, {'Col1': 1.064, 'Col2': 'e', 'Col3': ''}]
df = df.append(new_row, ignore_index=True)

def f(x):
    try:
        if x.is_integer():
            return f'{x: .0f}'
        else:
            return x
    except:
        return x

df = df.applymap(f)
print (df)

    Col1 Col2 Col3
0  1.150    c    2
1  1.064    e

Upvotes: 1

AlexNe
AlexNe

Reputation: 959

If you specifically want a string in your dataframe you cam use the try and ask for forgiveness pattern

def truncate(x):
    try:
        return '{0:1.0f}'.format(x)
    except ValueError:
        return x

Unfortunately I cannot test this directly on a dataframe.

Upvotes: 2

meTchaikovsky
meTchaikovsky

Reputation: 7666

Here is a simple solution

convert = lambda x: '%.0f' % x if isinstance(x,float) else x
df['Col3'] = df['Col3'].apply(convert)

Upvotes: 2

ZahraRezaei
ZahraRezaei

Reputation: 261

I suggest you this code:

df.astype(int)

Upvotes: 1

Related Questions