CodeGRU: Context-aware Deep Learning with Gated Recurrent Unit for Source Code Modeling

This repository contains the replication pack for our work "CodeGRU: Context-aware Deep Learning with Gated Recurrent Unit for Source Code Modeling"

Prerequisites

Make sure your system meets the followings requirements

• Python 3.6+
• Tensorflow 1.13.1+
• Keras 2.2.2+

Reproduction

Download the following files for reproduction.

• Vectors: Project vectors to replicate the results.
• SavedModels: Trained models to replicate the results.

Note:

• Use the Encoded (fixed size context) vectores to evaluate GRU models, use Encoded-VSCL (Variable size context) to evalute CodeGRU models. Both, Vectors and Saved Models can be downloaded here (Vectors,SavedModels).

Citation

if you use CodeGRU in academic works, please use the following citation:

@article{Hussain2020CodeGRU,
author    = {Yasir Hussain and
            Zhiqiu Huang and
            Yu Zhou and
            Senzhang Wang
            },
title     = {CodeGRU: Context-aware Deep Learning with Gated Recurrent Unit for
            Source Code Modeling},
journal   = {Information and Software Technology},
volume    = {Volume 125, September 2020, 106309},
year      = {2020},
url       = {https://doi.org/10.1016/j.infsof.2020.106309},
eprint    = {http://arxiv.org/abs/1903.00884}
}