The Quora Question Pairs dataset is used in this paper.
For a proper comparison to the published S-BERT models, the accompanying dataset splits were also used. These can be found here. The splits used are in the classification within the zip file, and should be placed in the data/sbert/ directory of this project.
The Python dependencies are listed in the requirements.txt file. You can install them by running the following command from the src directory:
pip install -r requirements.txt