This code achieves fine-tuning of DistilBERT model on C++ training set of vulnerable/non-vulnerable code.
As I got error message: "TypeError: new(): invalid data type 'str'" when trying to validate the model, this functionality from original codebase:
https://colab.research.google.com/github/DhavalTaunk08/Transformers_scripts/blob/master/Transformers_multilabel_distilbert.ipynb#scrollTo=kT5-oqMPB6vp
Has been modified/commented out.
A separate file is available with the non-modified/commented out validation functionality is available, should a member of the community wish to take it upon themselves to get the validation functionality to run.