Advancing Hate Speech Detection with Transformers: Experiments and Insights from the MetaHate Dataset

Fig: Encoder-Based (Left) and Decoder-Based (Right) Architectures of the Transformer

We used a Linux server equipped with an NVIDIA GPU, with following required packages:
cuda=12.1
python=3.12.4
pytorch==2.4.0
scikit-learn=1.2.2
transformers=4.44.0

The coding reference for this project are from
https://github.com/zekaouinoureddine/Opinion-Transformers

Step1: Dataset can be requested @ https://huggingface.co/datasets/irlab-udc/metahate
Step2: Run python3 file_name.py to run the model.