Downloads · 30 days
3
1% of all-time downloads
MichaelHarborg/NMT_da-en_translator
NMT_da-en_translator is a machine learning model from MichaelHarborg. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as mit.
Transformer model based on Vaswani et al., 2017 for Danish-English Neural Machine Translation.
Downloads · 30 days
3
1% of all-time downloads
All-time downloads
305
Public
Repo size
595 MB
Likes
0
Public
Click a slice to open those files.
.bin298 MB · 99%
From the Hugging Face model README
Transformer model based on Vaswani et al., 2017 for Danish-English Neural Machine Translation.
It has ~74M parameters and is a fine-tuned version of Helsinki-Opus-NLP da-en.
The model achieves a BLEU score of 49.16 on a hold-out test set for the TED2020 dataset (in-domain dataset).
The model achieves a BLEU score of 44.16 on a hold-out test set for the for CCAligned and Wikimatrix (out-of-domain dataset).
This outperforms the baseline Opus model, which achieved BLEU scores of 46.74 and 42.31 on the in-domain and out-of-domain data respectively.
Note: When running inference "_" characters can sometimes replace spaces.