lingua-py

lingua-py copied to clipboard

Support ONNX format for language models

4

Hi! I'm reaching out to kindly request the availability of the ONNX file for the language detector model that is currently being utilized within the project. Thanks.

BaMarcy

Would be interesting to have "prompt + LLM" accuracy in the benchmark as well. A simple prompt to GPT4 and restricting the output with LMQL should be quite straightforward.

ogencoglu

Confidence scores on multiple languages detection

1

Hi! Is there a specific method to get confidence scores when detecting multiple languages in mixed-language texts? Should I calculate it myself based on the DetectionResult occurrences?

Galvo87

Improve single language detection when words in other languages are quoted

2

When I put in german sentences with japanese words quoted then it might happen, that lingua claims it's 100% japanese. For example: `Wir stoßen an: "かんぱい". Er lächelte.` (in english,...

schrmh

Single word greeting detection issue

3

Based on the reported graphs I was expecting a high single-word detection accuracy, however when I tested some simple greetings, results were quite poor. I'm thinking that I might have...

AlexUmnov

Since the Language class and IsoCode classes are Rust classes, it would be nice to provide the `from_str` methods back so I can dynamically import iso codes.

lhr0909

Citation

1

Hi there, I am using your repo for my master's thesis. How I can cite it? Thanks! :)

ju-resplande

### Discussed in https://github.com/pemistahl/lingua-py/discussions/128 Originally posted by **thirtha-prasad** March 3, 2023 Getting below error, Package in installed using PIP and available in the python path. ImportError: cannot import name 'Language'...

asmita-mukherjee

Given a text in Ukrainian, two methods provide two completely different results. ```python detector = LanguageDetectorBuilder.from_all_languages().build() string = "Що найбільше подобається читачам у жанрі \"Фентезі\"?" print(detector.compute_language_confidence_values(string)) >>> [ConfidenceValue(language=Language.KAZAKH, value=1), ConfidenceValue(language=Language.AFRIKAANS,...

EvGe22

Accuracy comparision between lingua and fasttext

3

I just wanted to share these considering the comparisions mentioned in the readme. * I see that fasttext seems to be better overall for my usecase, though fasttext may just...

vanga