Ct2fast Opus Mt De En

🚀 Fast-Inference with Ctranslate2

This project speeds up inference by 2x - 8x using int8 inference in C++. It provides a quantized version of Helsinki-NLP/opus-mt-de-en, which is compatible with ctranslate2 and hf-hub-ctranslate2.

🚀 Quick Start

📦 Installation

To get started, you need to install the necessary packages:

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

💻 Usage Examples

Basic Usage

First, convert the model using the following command:

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

Then, you can use the model for inference:

from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer

model_name = "michaelfeil/ct2fast-opus-mt-de-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-de-en")
)
outputs = model.generate(
    text=["How do you call a fast Flan-ingo?", "User: How are you doing?"],
)
print(outputs)

Compatibility

• compute_type=int8_float16 for device="cuda"
• compute_type=int8 for device="cpu"

📚 Documentation

Original description

opus-mt-de-en

• Source languages: de
• Target languages: en
• OPUS readme: de-en
• Dataset: opus
• Model: transformer-align
• Pre-processing: normalization + SentencePiece
• Download original weights: opus-2020-02-26.zip
• Test set translations: opus-2020-02-26.test.txt
• Test set scores: opus-2020-02-26.eval.txt

Benchmarks

📄 License

This is just a quantized version. License conditions are intended to be identical to the original Hugging Face repo. The license used is Apache-2.0.