Performs Code Summarization, Bug Detection, Bug Removal using different Natural language processing models including Garph CodeBERT, GREAT, GNN, CoText etc.
The python notebooks have dataset and pre-trained model links. Following is the list of models used and their respective tasks performed.
| Models | Task | Result |
|---|---|---|
| Code2Seq | Code Summarization | Precision: 0.6 |
| Code2Vec | Method Name Prediction | Precision: 0.49 |
| GraphCodeBERT | Clone Detection | Blue score: 53.62 |
| GraphCodeBERT | Bug Repair | F1 score: 0.75 |
| GraphCodeBERT | Var Misuse | Precision: 0.66 |
| CoTexT | Bug Repair | Accuracy: 1.0 |
| GINN | Var Misuse | Precision: 0.76 |
| GREAT | Var Misuse | Accuracy: 89.01% |