|
--- |
|
license: apache-2.0 |
|
base_model: Helsinki-NLP/opus-mt-en-ru |
|
tags: |
|
- translation |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: ml_translation_model1 |
|
results: [] |
|
language: |
|
- en |
|
- ru |
|
datasets: |
|
- glazzova/tico19_en_ru |
|
pipeline_tag: translation |
|
--- |
|
|
|
|
|
# translation_en_ru |
|
|
|
Эта модель дообучена на [Helsinki-NLP/opus-mt-en-ru](https://huggingface.co/Helsinki-NLP/opus-mt-en-ru) с помощью |
|
датасета tico19, который содержит терминологию связанную с COVID-19. Модель может использоваться для перевода медицинских |
|
текстов |
|
|
|
|
|
## Результаты на тестовых данных: |
|
- Loss: 1.1217 |
|
- Bleu: 30.84 |
|
|
|
### Запуск модели |
|
```python |
|
from transformers import pipeline |
|
|
|
model_checkpoint = "glazzova/ml_translation_model1" |
|
translator = pipeline("translation", model=model_checkpoint) |
|
translator("i have a little cold and a cough") |
|
|
|
# у меня есть простуда и кашель |
|
``` |
|
|
|
### Гиперпараметры |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.19.1 |
