Usage suggestions for noobs

[sdspieg]

Could you please provide noob-proof instructions for how to use lingua to detect and language(s) used in a text column of a dataframe?
I installed it with pip install lingua-language-detector, which seemed to work just fine.
But when I then tried

import os
import pandas as pd
from lingua import LanguageDetectorBuilder

languages = ['en', 'fr', 'de', 'es']
detector_builder = LanguageDetectorBuilder.from_languages(*languages)

LanguageDetectorBuilder object
def detect_language(text):
detector = detector_builder.build()
return detector.detect(text).language.name

df['language'] = df['abstract'].apply(detect_language)

it threw me an error:

ImportError Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_58052\2805698229.py in
1 import os
2 import pandas as pd
----> 3 from lingua import LanguageDetectorBuilder
4
5 # Initialize the LanguageDetectorBuilder object with the languages to be supported

ImportError: cannot import name 'LanguageDetectorBuilder' from 'lingua' (c:\ProgramData\Anaconda3\lib\site-packages\lingua_init_.py)

[pemistahl]

Your error has nothing to do with pandas. You are trying to pass a list of language iso code strings to LanguageDetectorBuilder. But that's not how the API works. Please look into section 9 of the readme. You need to pass a list of Language enum instances.