Eye-tracking based classification of Mandarin Chinese readers with and without dyslexia using neural sequence models

This repository contains the sequence and baseline models used in Eye-tracking based classification of Mandarin Chinese readers with and without dyslexia using neural sequence models.

Using the models

Create a conda environment with

$ conda env create -f environment.yml

Then activate the environment and install your appropriate version of PyTorch.

$ conda install -y pytorch torchvision cudatoolkit=11.1 -c pytorch
$ # conda install pytorch torchvision cpuonly -c pytorch
$ pip install datasets transformers

Citation

Patrick Haller, Andreas Säuberli, Sarah Kiener, Jinger Pan, Ming Yan, and Lena Jäger. 2022. Eye-tracking based classification of Mandarin Chinese readers with and without dyslexia using neural sequence models. In Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022), pages 111–118, Abu Dhabi, United Arab Emirates (Virtual). Association for Computational Linguistics.