jasonhoang6201/primevul-for-linevul

original dataset: https://huggingface.co/datasets/colin/PrimeVul this dataset is created by: filter out 26% records with func > 512 tokens filter project record has < 2 samples duplicate vul records and split dataset into train, val, test to match distribute ratio in BigVul dataset