This directory stores the dataset inputs and generated parquet files used by the PhoBERT notebook in train/.
The dataset directory is ignored by git except this README. Keep model packages in train/artifacts/, not here.
The materialized dataset is hosted on Hugging Face:
The upstream GitHub… See the full description on the dataset page:
https://huggingface.co/datasets/dathuynh1108/vietnamnet-news.