In [2]:

    

# there seems to be '-' in some fields instead of NAs...
# let's remove the carets and check if the datatype of the column changed...
import pandas as pd
import numpy as np

df = pd.read_csv("input/input.csv")

df.head()


    

    
Out[2]:






  

    

      

      
Action_id
      
TypeAction
      
Page_name
      
Feed_message
      
User_name
      
CommentsForItem_message
      
Sentiment
      
Action_created_time_qvdate
    
  
  

    

      
0
      
Comment/1364017066952510_1364097753611108
      
Comment
      
Resident Evil
      
[UK ONLY] To celebrate the release of Resident...
      
Ronald Briand
      
Muy buena :)  (Y)
      
0.9880
      
1/5/2017
    
    

      
1
      
Comment/10154797365350629_10154797373500629
      
Comment
      
Bandai Namco
      
We continue our Tales Of look back with Tales ...
      
Jon Imboden
      
I may be in the minority here, but I LOVE the ...
      
0.9540
      
12/27/2016
    
    

      
2
      
Comment/1365217000165850_1365754593445424
      
Comment
      
Resident Evil
      
Are you afraid of the dark? Because you should...
      
Claudia Alejandra Olguín
      
mira Roberto Zeballos,  para probar en tu PEQU...
      
0.9527
      
1/7/2017
    
    

      
3
      
Comment/1364017066952510_1364061950281355
      
Comment
      
Resident Evil
      
[UK ONLY] To celebrate the release of Resident...
      
David Lawrence
      
Sarah Amadeus :O
      
0.9402
      
1/5/2017
    
    

      
4
      
Comment/1365217000165850_1225502404193085
      
Comment
      
Resident Evil
      
Are you afraid of the dark? Because you should...
      
Leandro Guilherme
      
Aeee Paulo Lucas vou te dar de presente no seu...
      
0.9382
      
1/8/2017
    
  

In [10]:

    

for column in df.columns.values:
    df.loc[df[column] == '-', column] = np.nan

df.head()


    

    
Out[10]:






  

    

      

      
Action_id
      
TypeAction
      
Page_name
      
Feed_message
      
User_name
      
CommentsForItem_message
      
Sentiment
      
Action_created_time_qvdate
    
  
  

    

      
0
      
Comment/1364017066952510_1364097753611108
      
Comment
      
Resident Evil
      
[UK ONLY] To celebrate the release of Resident...
      
Ronald Briand
      
Muy buena :)  (Y)
      
0.9880
      
1/5/2017
    
    

      
1
      
Comment/10154797365350629_10154797373500629
      
Comment
      
Bandai Namco
      
We continue our Tales Of look back with Tales ...
      
Jon Imboden
      
I may be in the minority here, but I LOVE the ...
      
0.9540
      
12/27/2016
    
    

      
2
      
Comment/1365217000165850_1365754593445424
      
Comment
      
Resident Evil
      
Are you afraid of the dark? Because you should...
      
Claudia Alejandra Olguín
      
mira Roberto Zeballos,  para probar en tu PEQU...
      
0.9527
      
1/7/2017
    
    

      
3
      
Comment/1364017066952510_1364061950281355
      
Comment
      
Resident Evil
      
[UK ONLY] To celebrate the release of Resident...
      
David Lawrence
      
Sarah Amadeus :O
      
0.9402
      
1/5/2017
    
    

      
4
      
Comment/1365217000165850_1225502404193085
      
Comment
      
Resident Evil
      
Are you afraid of the dark? Because you should...
      
Leandro Guilherme
      
Aeee Paulo Lucas vou te dar de presente no seu...
      
0.9382
      
1/8/2017
    
  

In [11]:

    

df.to_csv("output/data_cleaned.csv")