CoolFace
Modelpublic

michaelfeil/ct2fast-opus-mt-ROMANCE-en

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face
1likes35downloads
Model Card

# Fast-Inference with Ctranslate2

Speedup inference by 2x-8x using int8 inference in C++

quantized version of Helsinki-NLP/opus-mt-ROMANCE-en

bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0

Converted using

ct2-transformers-converter --model Helsinki-NLP/opus-mt-ROMANCE-en --output_dir /home/michael/tmp-ct2fast-opus-mt-ROMANCE-en --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16

Checkpoint compatible to ctranslate2 and hf-hub-ctranslate2

  • —compute_type=int8_float16 for device="cuda"
  • —compute_type=int8 for device="cpu"
python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer

model_name = "michaelfeil/ct2fast-opus-mt-ROMANCE-en"
# use either TranslatorCT2fromHfHub or GeneratorCT2fromHfHub here, depending on model.
model = TranslatorCT2fromHfHub(
        # load in int8 on CUDA
        model_name_or_path=model_name, 
        device="cuda",
        compute_type="int8_float16",
        tokenizer=AutoTokenizer.from_pretrained("Helsinki-NLP/opus-mt-ROMANCE-en")
)
outputs = model.generate(
    text=["How do you call a fast Flan-ingo?", "User: How are you doing?"],
)
print(outputs)

Licence and other remarks:

This is just a quantized version. Licence conditions are intended to be idential to original huggingface repo.

Original description

opus-mt-ROMANCE-en

Benchmarks

testsetBLEUchr-F
Tatoeba.fr.en62.20.750