Hengam: An Adversarially Trained Transformer for Persian Temporal Tagging

HuggingFace

• HengamTrans Space
• HengamTagger (Parstdex) Space
• HengamTrans Models (ModelCard)
• HengamCorpus (Dataset)

Code (Software)

HengamTagger

The HengamTagger is distributed as Parstdex package (will be continuously updated) and is available via pip.

HengamTransformer

In this directory 4 different notebooks are provided. Click on the hyperlinks to open each in Google Colab.

• Inference and Test
  • Inference_HengamTransW.ipynb: In this notebook, HengamTransW.pth is downloaded from Hengam HuggingFace model card, and an inference is provided in the last cells.
  • Inference_HengamTransA.ipynb: In this notebook, HengamTransA.pth is downloaded from Hengam HuggingFace model card, and an inference is provided in the last cells.
• Train
  • Train_HengamTransW.ipynb: This notebook uses the HengamCorpus dataset uploaded on HuggingFace. Afterwards, the HengamTransW will be trained.
  • Train_HengamTransA.ipynb: This notebook uses the trained HengamTransW.pth downloaded from HuggingFace and then trains it on strong labeled data in an adverserial manner to produce HengamTransA.

Data

Train Data

HengamCorpus data along with strong labeled data is uploaded in HuggingFace. Click on hyperlinks to download.

• HengamCorpus
  • HengamCorpus train data
  • HengamCorpus test data
  • HengamCorpus validation data
• Strong labeled data

Evaluation Data

HengamGold, challenge set and prediction result of different models on HengamGold is available in evaluation folder.

Citation

If you use any part of this repository in your research, please cite it using the following BibTex entry.

@inproceedings{mirzababaei-etal-2022-hengam,
	title        = {Hengam: An Adversarially Trained Transformer for {P}ersian Temporal Tagging},
	author       = {Mirzababaei, Sajad  and Kargaran, Amir Hossein  and Sch{\"u}tze, Hinrich  and Asgari, Ehsaneddin},
	year         = 2022,
	booktitle    = {Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing},
	publisher    = {Association for Computational Linguistics},
	address      = {Online only},
	pages        = {1013--1024},
	url          = {https://aclanthology.org/2022.aacl-main.74}
}