OpenNMT: Open-Source Toolkit for Neural Machine Translation
About
We describe an open-source toolkit for neural machine translation (NMT). The toolkit prioritizes efficiency, modularity, and extensibility with the goal of supporting NMT research into model architectures, feature representations, and source modalities, while maintaining competitive performance and reasonable training requirements. The toolkit consists of modeling and translation support, as well as detailed pedagogical documentation about the underlying techniques.
Guillaume Klein, Yoon Kim, Yuntian Deng, Jean Senellart, Alexander M. Rush• 2017
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Abstractive Summarization | Gigaword (test) | ROUGE-136.73 | 58 | |
| Formality Style Transfer | GYAFC Family & Relationships 1.0 (test) | BLEU0.403 | 15 | |
| Formality Style Transfer | GYAFC Entertainment & Music 1.0 (test) | BLEURT-0.919 | 15 | |
| Abstractive Summarization | Gigaword non-anonymized (test) | ROUGE-136.73 | 7 | |
| Word substitution decipherment | Yelp | BLEU (20% Sub)99 | 6 | |
| Summarization | Gigaword 100 randomly-selected articles (test) | Informativity3.26 | 6 | |
| Word order recovery | Yelp | BLEU64.6 | 5 |
Showing 7 of 7 rows