Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
liyankang 2a8c8f543d | 2 years ago | |
---|---|---|
scripts | 2 years ago | |
thseq | 2 years ago | |
.gitignore | 2 years ago | |
LICENSE | 2 years ago | |
README.md | 2 years ago | |
abdtrain.py | 2 years ago | |
mulsrctrain.py | 2 years ago | |
options.py | 2 years ago | |
requirements.txt | 2 years ago | |
setup.py | 2 years ago | |
state.py | 2 years ago | |
train.py | 2 years ago | |
translate.py | 2 years ago |
Zhang, Xiangwen, Jinsong Su, Yue Qin, Yang Liu, Rongrong Ji, and Hongji Wang. "Asynchronous bidirectional decoding for neural machine translation." In Thirty-Second AAAI Conference on Artificial Intelligence. 2018.
Python 3.7+ is required.
Install requirements as follows.
pip install -r requirements.txt
CUDA_VISIBLE_DEVICES=0 python -u abdtrain.py --train /path/to/train.en /path/to/train.de.pseudo.r2l /path/to/train.de --dev /path/to/dev.en /path/to/dev.de.pseudo.r2l /path/to/dev.de --vocab /path/to/vocab.en /path/to/vocab.de --vocab-size 99999 --model trained/run --max-step 200000 --eval-steps 5000 --shuffle 1 --batch-size 3000 3000 --batch-by-sentence 0 --arch abdrnn --criterion smoothed-ce --lr-scheduler noam --model-size 512 --warmup-steps 8000 --lr 1 --weight-tying 1 --accumulate 2 --ver 1
python -u translate.py --models trained/run -b 1 \
--input /path/to/test.en /path/to/test.de.pseudo.r2l > /path/to/test.de.mt
@inproceedings{zhang2018asynchronous,
title={Asynchronous bidirectional decoding for neural machine translation},
author={Zhang, Xiangwen and Su, Jinsong and Qin, Yue and Liu, Yang and Ji, Rongrong and Wang, Hongji},
booktitle={Thirty-Second AAAI Conference on Artificial Intelligence},
year={2018}
}
Zhang, Xiangwen, Jinsong Su, Yue Qin, Yang Liu, Rongrong Ji, and Hongji Wang. "Asynchronous bidirectional decoding for neural machine translation." In Thirty-Second AAAI Conference on Artificial Intelligence. 2018.
Python
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》