This repository contains code and data used in the paper "A Girl Has A Name, And It’s ... Adversarial Authorship Attribution for Deobfuscation" accepted at ACL 2022.
- Install Python3
- Download trained models and data for training and testing in the main folder.
- Install libraries needed for the project through
pip3 install -r requirements.txt
To evaluate directly using previous data, run the following in the main folder.
python3 eval.py
To evaluate the METEOR score from previous data, run the following in the main folder.
python3 CalcMeteor.py