Distil-lang-detect is text language detection module based on sequence classification technique DistilBERT by 🤗 Huggingface Transformers.
Distil-Lang-Detect can be easily fired-up. Just need to the following.
- python 3.5
- torch >= 1.2.0
- transformers >= 2.2.2
git clone https://github.com/yash1994/distil-lang-detect.git
cd dframcy
python setup.py install
from distillangdetect.detector import Detector
dct = Detector(device="cpu")
det = dct.detect("I love retro computing.")
print(det)
'English'
- Extensive testing.
- Add training and evaluation scripts.
- Output format options.
- Batch Processing.
- Bechmarking on different datasets.