CODEWITHSUNDEEP
pandas
shantanuo
shantanuo

Reputation: 32296

Apply a function data frame column

I have a data frame where 4th column contain ASCII values.

mydict=[('1385145553847005',
  '54',
  'NPAVIN',
  '9175042231172',
  '[89,111,117,114,32,78,80,65,86,32,79,110]',
  '20131123000914',
  'NA',
  'NA',
  '0',
  '0',
  'NA',
  'undefined',
  'a4d05539-cd61-43ee-a870-702e20caeaff',
  '0',
  '0',
  '0'),
('1385145553847006',
  '55',
  'NPAVIN1',
  '9175042231171',
  '[78,80,65,86,32,79,110]',
  '20131123000915',
  'NA',
  'NA',
  '0',
  '0',
  'NA',
  'undefined',
  'a4d05539-cd61-43ee-a870-702e20caeaff',
  '0',
  '0',
  '0')
] 

import pandas as pd
df = pd.DataFrame(mydict)

I have written a function that needs to be applied to the 4th column.

def get_ascii(amyl):
    mys=''
    for item in amyl:
        mys= mys+(chr(int(item)))
    return mys

This does not work. I get value Error:

df.apply(get_ascii(df[4]))

Expected Result: The following 2 values should be added to the last column of the data-frame.

'Your NPAV On'
'NPAV On'

Upvotes: 0

Views: 33

Answers (1)

EdChum
EdChum

Reputation: 393863

There are a number of problems here, your values is a literal string of a list of values: 

In [295]:

df[4]
Out[295]:
0    [89,111,117,114,32,78,80,65,86,32,79,110]
1                      [78,80,65,86,32,79,110]
Name: 4, dtype: object

I don't know if this is intended or not but you have to convert this to a list.

Secondly your function is working on the entire Series but it's expecting a single row value at a time, if you want to work a row at a time you have to pass pass param axis=1.

Anyway the following works by evaluating the series that is passed in, accessing the only element value in that series, evaluating as a list and then performing the transliteration:

In [294]:

import ast
def get_ascii(amyl):
    l = ast.literal_eval(amyl.values[0])
    mys=''
    for item in l:
        mys= mys+(chr(item))           
    return mys
df[[4]].apply(get_ascii, axis=1)
Out[294]:
0    Your NPAV On
1         NPAV On
dtype: object

Upvotes: 1

Related Questions