Instructions to use Devsyril/opus-mt-ee-fr-onnx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use Devsyril/opus-mt-ee-fr-onnx with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('translation', 'Devsyril/opus-mt-ee-fr-onnx');
Devsyril/opus-mt-ee-fr-onnx
Export ONNX du modèle Helsinki-NLP/opus-mt-ee-fr (Helsinki-NLP / OPUS-MT),
généré avec 🤗 Optimum.
Contient les composants ONNX standards pour un modèle seq2seq :
encoder_model.onnx, decoder_model.onnx, decoder_with_past_model.onnx (selon la version d'Optimum utilisée),
ainsi que le tokenizer d'origine.
Utilisation
from optimum.onnxruntime import ORTModelForSeq2SeqLM
from transformers import AutoTokenizer
repo_id = "Devsyril/opus-mt-ee-fr-onnx"
model = ORTModelForSeq2SeqLM.from_pretrained(repo_id)
tokenizer = AutoTokenizer.from_pretrained(repo_id)
inputs = tokenizer("Ton texte ici", return_tensors="pt")
generated = model.generate(**inputs)
print(tokenizer.batch_decode(generated, skip_special_tokens=True))
Modèle source : Helsinki-NLP/opus-mt-ee-fr
- Downloads last month
- -