Mocktail++

• A novel source code representation for capturing syntactic and semantic program features by utilizing multiple intermediate representations of code.

• Previous Versions

  • D. Vagavolu, K. C. Swarna and S. Chimalakonda, "A Mocktail of Source Code Representations," 2021 36th IEEE/ACM International Conference on Automated Software Engineering (ASE), 2021, pp. 1296-1300, doi: 10.1109/ASE51524.2021.9678551. [IEEE Xplore]

  • Dheeraj Vagavolu, Karthik Chandra Swarna, and Sridhar Chimalakonda. "A Mocktail of Source Code Representations." arXiv preprint arXiv:2106.10918 (2021) [arXiv]

• Please find the README for both the Path Extractor tool and the Extended code2vec model in their respective folders.

Datasets

Method Naming

• Download and create raw datasets using this script.
• Download the processed datasets from here. Processed datasets are path-extracted versions of the raw datasets. They can be readily used to train the model.

Program Classification and Clone Detection

• Download the raw datasets from CodeXGlue's repository.
• Download the processed datasets from here. Processed datasets are path-extracted versions of the raw datasets. They can be readily used to train the model.

Team

• In case of any queries or if you would like to give any suggestions, please feel free to contact:
  • Karthik Chandra (cs17b026@iittp.ac.in)
  • Noble Saji Mathews (ch19b023@iittp.ac.in)
  • Dheeraj Vagavolu (cs17b028@iittp.ac.in)
  • Sridhar Chimalakonda (ch@iittp.ac.in)