CODEWITHSUNDEEP
pythonpandas
Michele Montaruli
Michele Montaruli

Reputation: 25

Pandas apply a function that return a list to more column

I've a function that perform a modification on a string, under some condition than return a list containing modified string and a boolean check if new string is obtained. I want apply func to pandas dataframe column and store result in two new created columns. I found an inelegant way to achieve this purpose:

The main method is:

def alter_string(astring):
  ...
  return altered_string, boolean_check

def _perform_mod(astring):
  return alter_string(astring)[0]

def _check():
  return alter_string(astring)[1]

df['modified']=df['original'].apply(_perform_mod)
df['check']=df['original'].apply(_check)

In this way I achieve my goal but I have to run twice an heavy computational method. I wonder if there is a better way

Add some details to clarify my question I've a dataframe column 'original_string' containing string that are molecular descriptors, I apply to this string a function that can modify or not the string under some circumstances. The function return modified string and True or same string and False. I need to add two new columns to dataframe, modified_string and check Here a short sample

original --->  modified check

AAAAAA ----->  AAAAAA False
AAABCD ----->  AAAVCD True
ACCBDE ----->  AACADE True

`

Upvotes: 1

Views: 135

Answers (1)

Teoretic
Teoretic

Reputation: 2533

Try this:

df['modified'], df['check'] =  zip(*df['original'].apply(alter_string))

This way you run alter_string function only once.

zip function creates a list of tuples, where each tuple is a series. Then through tuple unpacking (df['modified'], df['check']) we create our brand new columns in dataframe.

Based on this answer

Upvotes: 1

Related Questions