Code and Data for paper: Parsing Natural-Language based Software Requirements into Code Segments
All the data files are placed under data folder:
File | Mode | Use |
---|---|---|
Reqset-train.csv | Plain Text | Parse to generate training dataset |
Reqset-test.csv | Plain Text | Parse to generate testing dataset |
Reqset-train-tagged.csv | Parsed | Dataset to use for training |
Reqset-test-tagged.csv | Parsed | Dataset to use for testing |
Reqset-comb-tagged.csv | Parsed | Dataset to split for training and testing |
Code is placed inder code folder. There are two notebooks:
- ReqCRF.ipynb: Contains code for BILSTM-CRF model
- ReqBERT.ipynb: Contains code for fine-tuned BERT model
For any questions or comments reach out to: [email protected]