DIfference between output of sublime 3 vs jupyter

user_7299245 0
Viewed 232

I have a problem with the output of sublime text. When I run this:

import pandas as pd
import numpy as np


df = pd.DataFrame()
df['name'] = ['John', 'Steve', 'Sarah']
df.assign (age =[30,31,32])
print (df)

I get this result:

        name
    0   John
    1  Steve
    2  Sarah

I think this is a mistake, because I should can see the age column.

When I Run this code in jupyter I obtain:

    name    age
0   John    30
1   Steve   31
2   Sarah   32

Somebody have any idea what it happen?

thanks for the help

sublimetext3 python
2 Answers

This is far from being a mistake or an interpreter mismatch.

When you call df.assign(age =[30,31,32]), you are making a copy of df with an additional column age. When this is done in a REPL-like editor, you see the ouput rendered. But in most text editors, this will not show because you are not printing the ouput. You are using the __repr__ method of the DataFrame object. Unless the text editor has implemented a way to render __repr__ ouputs, you will never see the output of df.assign(age =[30,31,32]).

Now, jupyter, on the other hand, has a REPL feature, and therefore will show commands based on the output's __repr__ method, without needing to call the print function. As a result, jupyter will show the output of df.assign(age =[30,31,32]).

The bottom line remains that you are not changing the dataframe in place. Both sublime and jupyter would print the results, if the code was written as follows:

import pandas as pd
import numpy as np


df = pd.DataFrame()
df['name'] = ['John', 'Steve', 'Sarah']
df = df.assign(age=[30,31,32])
print(df)

I hope this helps.

user_31354170

The assign() of dataframe returns a new object (a copy) with all the original columns in addition to the new ones - see pandas.DataFrame.assign.

So you have to assign it for getting its return value:

df = df.assign (age =[30,31,32])
user_70235900
Related