ValueError: could not convert string to float Using Python

Question

I have been looking to remove all the non numerical character, and convert the same to float type, But whenever any String character comes up , it gives the Error as "ValueError: could not convert string to float"

Please suggest how to resolve it .

Input File

col1           col2  

122.45         NaN
Ninety Five    3585/-
9987           178@#?
225 Nine       1983.86
Twelve         7363*

Output File

col1           col2  

122.45         NaN
NaN            3585
9987           178
225            1983.86
NaN            7363

Code i am using :

df[['col1','col2']] = df[['col1','col2']].replace('([^\d/.])', '', regex=True).astype(float)

Getting the Error:

ValueError: could not convert string to float

Try with the pattern '[^\d\.]' and add .replace('', np.nan) like so: df[['col1','col2']].replace('([^\d\.])', '', regex=True).replace('',np.nan).astype(float) — Whole Brain
– Whole Brain, Commented Jun 18, 2021 at 15:25
@WholeBrain - Gives the same Error when it comes to the row containing value "One Lakh Two Thousand Three hundred & Twenty Paise" — Manz
– Manz, Commented Jun 18, 2021 at 15:32

Will Da Silva · Accepted Answer · 2021-06-18 17:02:12Z

1

You need to use a raw string (with the r in front) for regex patterns, or double backslash (\\) escapes. Also you need \. to match literal . characters, not /.:

df[['col1', 'col2']] = df[['col1', 'col2']].replace('(-?[^\d\.])', '', regex=True).replace('', float('NaN')).astype(float)

edited Jun 18, 2021 at 17:02

answered Jun 18, 2021 at 15:31

Will Da Silva

7,1512 gold badges35 silver badges55 bronze badges

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

7 Comments

Manz Over a year ago

@Willdasilva - Using the above lines as well Gives the same Error when it comes to the row containing value "One Lakh Two Thousand Three hundred & Twenty Paise"

Manz Over a year ago

@AnuragDabas - When we have the all characters as Alpha in row values, For Eg. "One Lakh Two Thousand Three hundred & Twenty Paise" It still gives the same Error.

Anurag Dabas Over a year ago

@Manz so

df[['col1','col2']]=df[['col1','col2']].replace('([^\d\.])', '', regex=True).replace('',float("NaN"),regex=True).astype(float)

doesn't work?

Manz Over a year ago

@willdasilva - Any suggestion on if having the row value as "-122.45", Its not working in this case, how to resolve the same.

Manz Over a year ago

@AnuragDabas - Thanks for the Answer , Its working but when we have the row values as "-122.45" it doesn't works , gives the same value as output.

|

Collectives™ on Stack Overflow

ValueError: could not convert string to float Using Python

1 Answer 1

7 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

7 Comments

Your Answer

Sign up or log in

Post as a guest

Related