Difference while using apply method and vector method

Q&A DQ Courses
,
#1

Screen Link: https://app.dataquest.io/m/345/transforming-data-with-pandas/7/reshaping-data-with-the-melt-function

Your Code: I need to understand why these two codes are giving a different result
code 1

def percentages(col):
    div = col/happiness2015['Happiness Score']
    return round(div*100,2)

melt['Percentage']=melt[['value']].apply(percentages)

Code 2
melt['Percentage']=round(melt['value']/melt['Happiness Score']*100,2)

What I expected to happen:

I was expecting both of them to give same result but code1 is giving NaN for many rows

Other details:
DF details
main_cols = [‘Country’, ‘Region’, ‘Happiness Rank’, ‘Happiness Score’]
factors = [‘Economy’, ‘Family’, ‘Health’, ‘Freedom’, ‘Trust’, ‘Generosity’, ‘Dystopia Residual’]

melt = pd.melt(happiness2015,id_vars= main_cols, value_vars= factors)

#2

Hi @hora.amit

Welcome to our Dataquest Community.

Code 1-
Two mistakes in it.

First, You are dividing a single value from value with the whole column Happiness Score or all values from the ‘Happiness Score’ column one by one. But you need only one value from it with the corresponding index.

Second, You need to assign the axis parameter in the method apply to understand how values to send either row or column-wise. By default axis=0 i.e. row-wise.

Corrected Code 1-

def percentages(row):
    value=row['value']
    score=row['Happiness Score']
    div = value/score
    return round(div*100,2)

melt['Percentage']=melt.apply(percentages, axis=1)

I have edited it.

Code 2-
It looks correct, for better understanding you can add brackets.

melt["Percentage"] = round(((melt["value"]/melt["Happiness Score"]) * 100), 2)

I hope this helps :slightly_smiling_face: .

2 Likes
#3

use div = col/melt[‘Happiness Score’]
instead of using “div = col/happiness2015[‘Happiness Score’]”

That’s works fine

#4

Hello, what is the value column? The instructions say to convert it to a percentage but there is no value column in the happiness2015 dataframe, as far as I can tell.

#5

Hi @vroomvroom

Please, read the working of pd.melt() function and the slides again carefully. You will find your answer.

1 Like
#6

I found the answer, thanks.

1 Like