# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Helsinki-NLP/opus-mt-cs-de")
model = AutoModelForSeq2SeqLM.from_pretrained("Helsinki-NLP/opus-mt-cs-de", device_map="auto")Quick Links
opus-mt-cs-de
source languages: cs
target languages: de
OPUS readme: cs-de
dataset: opus
model: transformer-align
pre-processing: normalization + SentencePiece
download original weights: opus-2020-01-20.zip
test set translations: opus-2020-01-20.test.txt
test set scores: opus-2020-01-20.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| newssyscomb2009.cs.de | 22.0 | 0.525 |
| news-test2008.cs.de | 21.1 | 0.520 |
| newstest2009.cs.de | 22.2 | 0.525 |
| newstest2010.cs.de | 22.1 | 0.527 |
| newstest2011.cs.de | 21.6 | 0.515 |
| newstest2012.cs.de | 22.2 | 0.516 |
| newstest2013.cs.de | 24.8 | 0.538 |
| newstest2019-csde.cs.de | 23.6 | 0.530 |
| Tatoeba.cs.de | 51.6 | 0.687 |
- Downloads last month
- 134
# 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-cs-de")