Convert column object into integer in python

Question

I have this dataset:

training.head()
Out[115]: 
GridID        date      Shift  Accident  
0       1  2010-10-08      Night         0                
1       1  2011-02-16    Morning         0                     
2       1  2014-05-31      Night         0                   
3       1  2011-04-03  Afternoon         0                   
4       1  2013-02-20    Morning         0

I would like to replace in column Shift the words "Morning", "Afternoon" and "Night" with integers 1, 2 and 3, repectively.

I tried:

training['Shift'].str.replace('Morninig','1').astype(int)

But it gives me:

 ValueError: invalid literal for int() with base 10: 'Night'

first replace all text and later change column to int using astype(int) — furas
– furas, Commented Dec 26, 2016 at 7:40
I tried to replace all texts but it gives me the error as shown above — Claudio
– Claudio, Commented Dec 26, 2016 at 7:43
it seams you didn't change all text - you still have some Night in column. Column can have only one type so astype(int) convert all elements in column. — furas
– furas, Commented Dec 26, 2016 at 7:47

DeepSpace · Accepted Answer · 2016-12-26 07:50:19Z

3

Use Series.replace instead of str.replace, and pass it a dictionary from old value to new value. Also make sure you use inplace=True or reassign it to the series.

import pandas as pd

df = pd.DataFrame({'a': ['morning', 'afternoon']})

print(df)

>>            a
   0    morning
   1  afternoon

df['a'].replace({'morning': 1,
                 'afternoon': 2}, inplace=True)

print(df)

>>    a
   0  1
   1  2

print(df['a'].dtype)

>> int64

answered Dec 26, 2016 at 7:50

DeepSpace

82.1k12 gold badges119 silver badges166 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Mohammad Yusuf · Accepted Answer · 2016-12-26 07:54:19Z

1

Try this:

df2=pd.Series(data={'Night':3, 'Morning':1, 'Afternoon':2})
df.Shift = df.Shift.map(df2)

df2 = pd.read_clipboard()
df3=pd.Series(data={'Night':3, 'Morning':1, 'Afternoon':2})
df2.Shift = df2.Shift.map(df3)
df2

Output:

edited Dec 26, 2016 at 7:54

answered Dec 26, 2016 at 7:48

Mohammad Yusuf

17.1k12 gold badges60 silver badges87 bronze badges

Collectives™ on Stack Overflow

Convert column object into integer in python

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related