Translation
Transformers
PyTorch
Safetensors
marian
text2text-generation
opus-mt-tc
Eval Results (legacy)
Instructions to use Helsinki-NLP/opus-mt-tc-big-de-zle with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Helsinki-NLP/opus-mt-tc-big-de-zle 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="Helsinki-NLP/opus-mt-tc-big-de-zle")# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("Helsinki-NLP/opus-mt-tc-big-de-zle") model = AutoModelForSeq2SeqLM.from_pretrained("Helsinki-NLP/opus-mt-tc-big-de-zle", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| deu-bel flores101-dev 0.38483 9.3 997 23996 | |
| deu-rus flores101-dev 0.54342 26.4 997 22657 | |
| deu-ukr flores101-dev 0.52143 23.3 997 21841 | |
| deu-bel flores101-devtest 0.39045 9.8 1012 24829 | |
| deu-rus flores101-devtest 0.54152 26.3 1012 23295 | |
| deu-ukr flores101-devtest 0.53286 24.2 1012 22810 | |
| deu-rus newstest2012 0.49409 20.8 3003 64790 | |
| deu-rus newstest2013 0.52631 24.9 3000 58560 | |
| deu-bel tatoeba-test-v2020-07-28 0.53161 29.5 550 3595 | |
| deu-rus tatoeba-test-v2020-07-28 0.67044 46.1 10000 67433 | |
| deu-ukr tatoeba-test-v2020-07-28 0.62698 40.8 10000 54213 | |
| deu-bel tatoeba-test-v2021-03-30 0.53247 29.5 554 3617 | |
| deu-rus tatoeba-test-v2021-03-30 0.66964 45.8 11265 76204 | |
| deu-ukr tatoeba-test-v2021-03-30 0.62795 40.8 10174 55335 | |
| deu-bel tatoeba-test-v2021-08-07 0.53128 29.5 551 3601 | |
| deu-rus tatoeba-test-v2021-08-07 0.67143 46.1 12800 87296 | |
| deu-ukr tatoeba-test-v2021-08-07 0.62737 40.7 10319 56287 | |