Can we do dataframe level exception handling for pandas?

Question

I am new to pandas, I want to know that does pandas dataframe have their own way of exception handling other than using try/ except python.

I have tried exec function of python to write entire try/except in one line but I want pandas specific syntax or way of exception handling that can be done in a single line.

Below is the code that I have tried:

import pandas as pd

import numpy as np

data1 = {'id'   : [1,2,3,4,5],
         'Rate' : [5,9,3,'A',6],
         'Name' : ['a','b','c','d','e']}

df = pd.DataFrame(data1)

df['M_Rate1'] = df['Rate'].apply(lambda x, y=exec("def f(s):\n try:\n  return int(s) * 2\n except ValueError as e:  return 'Only Number can be converted to int'"): f(x))

Is their a better way for exception handling in oneline in pandas?

This question is off-topic here. Can you provide a minimal reproducible example example of what you're doing? — IanS
– IanS, Commented Aug 5, 2019 at 7:31
It's this way I have tried single line exception handling. I am Hopping for better way of exception handling than above, that can be done for dataframe . — dhee
– dhee, Commented Aug 5, 2019 at 9:05
Why does it have to fit on one line? That seems like a very unnecessary requirement... — IanS
– IanS, Commented Aug 5, 2019 at 9:57
My actual code length is too long i.e. about 3000 line, so to reduce it code complexity i was looking for it. — dhee
– dhee, Commented Aug 5, 2019 at 10:24
Complexity is more about readability than actual lines of code. It also obviously involves other concepts, but that's out of context here. Please, read the Zen of Python -- and focus on these 2 sentences : "Sparse is better than dense. Readability counts." — IMCoins
– IMCoins, Commented Aug 5, 2019 at 11:32

IMCoins · Accepted Answer · 2019-08-05 09:50:11Z

2

It looks like you are trying to apply a custom function in a very bad way, using a lambda, with a function defined using eval within an optional parameter.

You should try and go for something like this:

import pandas as pd

data = {
    'id'   : [1,2,3,4,5],
    'Rate' : [5,9,3,'A',6],
    'Name' : ['a','b','c','d','e']
}
df = pd.DataFrame(data)

def f(x):
    try:
        return int(s) * 2
    except ValueError as e:
        return 'Only Number can be converted to int'

df['M_Rate1'] = df['Rate'].apply(f)
print(df)
#   id Rate Name                              M_Rate1
#0   1    5    a                                   10
#1   2    9    b                                   18
#2   3    3    c                                    6
#3   4    A    d  Only Number can be converted to int
#4   5    6    e                                   12

answered Aug 5, 2019 at 9:50

IMCoins

3,3161 gold badge13 silver badges27 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

dhee Over a year ago

thanks for your solution but i am looking for single line way to deal with the exception.

IanS · Accepted Answer · 2019-08-05 10:00:33Z

1

Use pandas to_numeric and coerce failed conversions:

df['M_Rate1'] = pd.to_numeric(df['Rate'], errors='coerce') * 2

And if you must have an error message (not recommended):

df['M_Rate1'] = pd.to_numeric(df['Rate'], errors='coerce').mul(2).fillna('error_message')

answered Aug 5, 2019 at 10:00

IanS

16.3k9 gold badges64 silver badges87 bronze badges

Collectives™ on Stack Overflow

Can we do dataframe level exception handling for pandas?

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related