Adding multiple columns of unique calculated values to a pandas dataframe

Question

I would like to add new columns to a data frame using function and values from the original data frame

Create the dataframe

df = pd.DataFrame({'f1'   : np.random.randn(10), 
                   'f2'   : np.random.randn(10),
                   'f3'   : np.random.randn(10),
                   'f4'   : np.random.randn(10),
                   'f5'   : np.random.randn(10)})

Test function to be applied to the existing columns

def testfun(x,n):
    return x * n

Arguments for the function - each new column has different arguments

colnum   = [1,2,3,4,5]

Create new columns names for the new columns to add to the data frame

newcol         = [s + "_D" for s in df.columns]

Loop through the existing columns applying the function and appropriate argument for that column. Each new column will be assigned an unique name.

This part function does not work!

for s in range(len(df.columns)):    
     df       = df.assign(newcol[s] = testfun(df[[df.columns[s]]], s))

The new data frame should contain 10 columns.

Clock Slave Clock Slave · Accepted Answer · 2017-08-24T03:13:35

You can try this

import pandas as pd
import numpy as np

df = pd.DataFrame({'f1'   : np.random.randn(10), 
                   'f2'   : np.random.randn(10),
                   'f3'   : np.random.randn(10),
                   'f4'   : np.random.randn(10),
                   'f5'   : np.random.randn(10)})

def testfun(x,n):
    return x * n

colnum = [1,2,3,4,5]
newcol = [s + "_D" for s in df.columns]
for s in range(len(df.columns)):    
    df.loc[:,newcol[s]] = df[[df.columns[s]]]*s

Adding multiple columns of unique calculated values to a pandas dataframe

2 Answers