Skip to content

Latest commit

 

History

History
18 lines (14 loc) · 494 Bytes

README.md

File metadata and controls

18 lines (14 loc) · 494 Bytes

LMGAN : Linguistically Informed Semi-Supervised GAN with Mixup for text classification

Run model

Following command will run the model.


$ bash run_lmgan.sh

Requirements

The reuqired packages and versions are described in reuqirements.txt

Datasets

You can find the AG's news, DBpedia, Yahoo! Answers datasets at following url : https://github.com/LC-John/Yahoo-Answers-Topic-Classification-Dataset

We only used body texts when using above datasets.