translate dataframe column with python

Question

i want to translate column of my dataframe from french to english.

 from googletrans import Translator 
 translator = Translator()         
 df = pd.DataFrame({'French':['ma voiture','Il fait beau aujourdhui']})
 df['English'] = df['French'].apply(translator.translate,src='fr',dest='en').apply(getattr, args=('text',))

I did these lines but i found an error.. How can i solve it please.?

Note that in line 1 you miss a comma after aujourd' and a quote before hui.Moreover, what is the definition of translator? — sentence
– sentence, Commented Apr 22, 2019 at 14:14
yepp .. from googletrans import Translator translator = Translator() — anayisse2
– anayisse2, Commented Apr 22, 2019 at 14:15
Unfortunately I do not use that package. As suggested by @chitown88, maybe you should try with other packages. — sentence
– sentence, Commented Apr 22, 2019 at 17:57

chitown88 · Accepted Answer · 2019-04-22 15:08:05Z

0

Well I was having an issue with googletrans. But then tried the solution here that suggests to do:

pip install googletrans-temp

And your code seemed to work just fine:

Output:

print (df)
                    French                         English
0               ma voiture                          my car
1  Il fait beau aujourdhui  Such a beautiful weather today

answered Apr 22, 2019 at 15:08

chitown88

29.1k6 gold badges34 silver badges67 bronze badges

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

10 Comments

anayisse2 Over a year ago

i have always the same problem of ** JSONDecodeError: Expecting value: line 1 column 1 (char 0)**

anayisse2 Over a year ago

same error do you see how to solve it .. I did all what you recommand

chitown88 Over a year ago

you get the same error with just the rows in the dataframe?

anayisse2 Over a year ago

yep with the same dataframe and the same rows

chitown88 Over a year ago

not really sure. May just have to find another library or package. I'm not too familiar with other ones. You might just have to search around and see.

|

Saurabh Khanna · Accepted Answer · 2022-10-08 22:32:33Z

0

Google Translate has limits on volume translated. EasyNMT is a scalable solution.

from easynmt import EasyNMT
model = EasyNMT("opus-mt")

df["English"] = df.apply(lambda row: model.translate(row["French"], target_lang="en"), axis=1)

answered Oct 8, 2022 at 22:32

Saurabh Khanna

5714 silver badges5 bronze badges

2 Comments

sifar Over a year ago

please specify a complete example. The above does not seem to work with EasyNMT, but gives

OSError: Helsinki-NLP/opus-mt-lt-en is not a local folder and is not a valid model identifier listed on 'https://huggingface.co/models' If this is a private repository, make sure to pass a token having permission to this repo either by logging in with

huggingface-cli login` or by passing `token=<your_token>``

Saurabh Khanna Over a year ago

Can you please share a data snapshot and the relevant code snippet you are running (helps understand the error)? I believe we are trying to translate from Lithuanian to English here. This is the relevant model on HuggingFace: huggingface.co/Helsinki-NLP/opus-mt-tc-big-lt-en

Collectives™ on Stack Overflow

translate dataframe column with python

2 Answers 2

10 Comments

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

10 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related