This code is TMU System for SLAM-2018
- python==3.6.0
- torch==0.4.1
- gensim==3.6.0
You need to download data from this site and place them to data directory. Preprocessing the data for model training:
$ mkdir model_data
$ python preprocessRNNLM.py -train_data data/en_es.slam.20171218.train -valid_data en_es.slam.20171218.dev -valid_key en_es.slam.20171218.dev.key -save_data model_data/en_es
Training the model:
$ python trainRNNLM.py -data model_data/en_es.train.pt -save_model model/enes -gpus 0
Testing the model:
$ python test.py -model model/[model name] -data data/en_es.slam.20171218.dev -output enes.pre
When referencing this code, please cite this paper.
@InProceedings{W18-0544,
author = "Kaneko, Masahiro
and Kajiwara, Tomoyuki
and Komachi, Mamoru",
title = "TMU System for SLAM-2018",
booktitle = "Proceedings of the Thirteenth Workshop on Innovative Use of NLP for Building Educational Applications",
year = "2018",
publisher = "Association for Computational Linguistics",
pages = "365--369",
location = "New Orleans, Louisiana",
url = "http://aclweb.org/anthology/W18-0544"
}