Change a value inside of a DataFrame in Python (pandas)-CodePudding

i wanna change all the cells inside my data frame that are string and value = ''.

I have one data set that has 7 columns . for example:

   a,b,c,d,e,f,g.

and has 700 rows. i wanna change the value of the cells in specific 5 columns in one code. I tried this:

columns = [a,b,c,d,e]

def get_tmp(i):
  if len(i) == 0:
    b ='tmp'
    return b
  else:
    return i


weights_df[colun] = weights_df[colun].apply(get_tmp)

but this don't function.

to fix the problem i used a looping for:

columns = [a,b,c,d,e]

def get_tmp(i):
  if len(i) == 0:
    b ='tmp'
    return b
  else:
    return i

for colun in columns:
  weights_df[colun] = weights_df[colun].apply(get_tmp)

have another way to fix this situation using only .apply? if have, Do i need change somethin in my function ? what i need change ?

thank you guys.

CodePudding user response：

You can try this code.

import pandas as pd
import numpy as np
filename = 'Book1.xlsx'
weights_df= pd.read_excel(filename)

columns = ['a','b','c','d','e']
for col in columns:
  weights_df[col] = 
  weights_df[col].apply(lambda x: 'tmp' if x=='' else x)

This code is working in my local.

CodePudding user response：

thank you Bright Gene, it's a good answer. in reality i wanna check if is possible to do the same changes with out the looping for.

I will be more clear, maybe i did some miscommunication.

this data frame have this columns a,b,c,d,e,f,g.

i wanna change only a party of these columns: a,b,c,d,e.

the other are numbers.

I created a list: columns_to_modify= [a,b,c,d,e]

I wanna try to change inside like this:

weights_df[columns_to_modify] = weights_df[columns_to_modify].apply(lambda x: 'tmp' if x=='' else x)

at this moment i wanna understand if have a way to apply only to a specific columns whit out lopping for.