Deep Code Search

A keras implementation of the paper Deep Code Search.

Dependency

Tested in Ubuntu 16.04

• Python 3.6
• Keras 2.3.1 or newer
• Tensorflow 2.0.0 or Theano 0.8.0~0.9.1

Code Structures

• models.py: Neural network models for code/desc representation and similarity measure.

• main.py: The main entry for code search, including four sub-tasks:

  1. Train: train the code/desc representaton models;
  2. Eval: evaluate the learnt code/desc representation models;
  3. Code Embedding: encode code into vectors and store them to a file;
  4. Search: search relevant code for a given query.

• configs.py: Configurations for models defined in the models.py. Each function defines the hyperparameters for the corresponding model.

Usage

Data Preparation

The /data folder provides a small dummy dataset for quick deployment. To train and test our model:

1. Download and unzip real dataset from Google Drive or Baidu Pan for Chinese users.

2. Replace each file in the /data folder with the corresponding real file.

Configuration

Edit hyper-parameters and settings in config.py

Train

python main.py --mode train

Code Embedding

First, set reload in config.py to the number of optimal checkpoint, e.g., 500

Then, run

python main.py --mode repr_code

Search

First, set reload in config.py to the number of optimal checkpoint, e.g., 500

Then, run

python main.py --mode search

Tool Demo

An online tool demo can be found at http://211.249.63.55:81/ (Unavailable Now)

Citation

If you find it useful and would like to cite it, the following would be appropriate:

@inproceedings{gu2018deepcs,
  title={Deep Code Search},
  author={Gu, Xiaodong and Zhang, Hongyu and Kim, Sunghun},
  booktitle={Proceedings of the 2018 40th International Conference on Software Engineering (ICSE 2018)},
  year={2018},
  organization={ACM}
}