Instructions to use NYTK/translation-bart-hu-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NYTK/translation-bart-hu-en with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("translation", model="NYTK/translation-bart-hu-en")# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("NYTK/translation-bart-hu-en") model = AutoModelForSeq2SeqLM.from_pretrained("NYTK/translation-bart-hu-en") - Notebooks
- Google Colab
- Kaggle
BART Translation model
For further models, scripts and details, see our repository or our demo site.
Source language: Hungarian
Target language: English
Pretrained on English WikiText-103 and Hungarian Wikipedia
Finetuned on subcorpora from OPUS
- Segments: 56.837.602
Limitations
- tokenized input text (tokenizer: HuSpaCy)
Results
| Model | BLEU | chrF-3 |
|---|---|---|
| Google en-hu | 25.30 | 54.08 |
| BART-base-enhu | 34.38 | 58.88 |
| Google hu-en | 34.48 | 59.59 |
| BART-base-huen | 38.03 | 61,37 |
Citation
If you use this model, please cite the following paper:
@inproceedings {yang-bart,
title = {{BARTerezzünk! - Messze, messze, messze a világtól, - BART kísérleti modellek magyar nyelvre}},
booktitle = {XVIII. Magyar Számítógépes Nyelvészeti Konferencia},
year = {2022},
publisher = {Szegedi Tudományegyetem, Informatikai Intézet},
address = {Szeged, Magyarország},
author = {Yang, Zijian Győző},
pages = {15--29}
}
- Downloads last month
- 79