Translation
Transformers
PyTorch
Russian
Uzbek
m2m_100
text2text-generation
nllb
russian
uzbek
Eval Results (legacy)
Instructions to use sarahai/nllb-ru-uz with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sarahai/nllb-ru-uz 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="sarahai/nllb-ru-uz")# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("sarahai/nllb-ru-uz") model = AutoModelForMultimodalLM.from_pretrained("sarahai/nllb-ru-uz") - Notebooks
- Google Colab
- Kaggle
| language: | |
| - ru | |
| - uz | |
| license: apache-2.0 | |
| library_name: transformers | |
| tags: | |
| - translation | |
| - nllb | |
| - russian | |
| - uzbek | |
| datasets: | |
| - custom_dataset | |
| widget: | |
| - text: >- | |
| src_lang="rus_Cyrl", tgt_lang="uzn_Latn" | |
| Не задавай глупых вопросов | |
| metrics: | |
| - bleu | |
| model-index: | |
| - name: NLLB-200-Distilled-600M Russian to Uzbek Translation | |
| results: | |
| - task: | |
| name: Text Translation | |
| type: text2text-generation | |
| dataset: | |
| name: "Custom Dataset (10k parallel sentences)" | |
| type: custom_dataset | |
| metrics: | |
| - name: BLEU Score | |
| type: bleu | |
| value: your_model's_bleu_score | |
| # Description in English | |
| # NLLB-200-Distilled-600M Russian to Uzbek Translation | |
| ## Model Description | |
| This model is a fine-tuned version of Meta's NLLB-200-Distilled-600M, tailored specifically for translating text from Russian to Uzbek. It leverages the powerful base of the NLLB framework, designed to provide high-quality translations across numerous languages, now with enhanced proficiency in the Russian to Uzbek language pair thanks to targeted fine-tuning. | |
| ### Python Code Example | |
| ```python | |
| from transformers import pipeline | |
| # Initialize the translation pipeline | |
| translator = pipeline("translation_ru_to_uz", model="sarahai/nllb-ru-uz") | |
| # Translate text from Russian to Uzbek | |
| russian_text = "Это пример предложения для перевода." | |
| translation = translator(russian_text) | |
| print("Translation:", translation[0]['translation_text']) | |
| ``` | |
| # Description in Russian | |
| # NLLB-200-Distilled-600M Перевод с русского на узбекский | |
| ## Описание Модели | |
| Эта модель является доработанной версией модели Meta NLLB-200-Distilled-600M, предназначенной специально для перевода текстов с русского на узбекский язык. Она использует мощную основу фреймворка NLLB, созданного для обеспечения качественного перевода между многими языками, теперь с улучшенной способностью в паре русский-узбекский благодаря целевой доработке. | |
| ### Пример кода на Python | |
| ```python | |
| from transformers import pipeline | |
| # Инициализация пайплайна перевода | |
| translator = pipeline("translation_ru_to_uz", model="sarahai/nllb-ru-uz") | |
| # Перевод текста с русского на узбекский | |
| russian_text = "Это пример предложения для перевода." | |
| translation = translator(russian_text) | |
| print("Перевод:", translation[0]['translation_text']) | |
| ``` | |
| # Планируется улучшение модели и совершенствование переводов путем тренировки на более объемном датасете. |