https://github.com/pcyin/nl2code
A syntactic neural model for parsing natural language to executable code
https://github.com/pcyin/nl2code
deep-learning programming-language
Last synced: 3 months ago
JSON representation
A syntactic neural model for parsing natural language to executable code
- Host: GitHub
- URL: https://github.com/pcyin/nl2code
- Owner: pcyin
- Created: 2017-04-20T17:33:19.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2022-11-12T01:12:53.000Z (about 3 years ago)
- Last Synced: 2025-10-10T16:32:45.461Z (3 months ago)
- Topics: deep-learning, programming-language
- Language: Python
- Size: 142 KB
- Stars: 186
- Watchers: 9
- Forks: 33
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# NL2code
A syntactic neural model for parsing natural language to executable code [paper](https://arxiv.org/abs/1704.01696).
## Dataset and Trained Models
Get serialized datasets and trained models from [here](https://drive.google.com/drive/folders/0B14lJ2VVvtmJWEQ5RlFjQUY2Vzg). Put `models/` and `data/` folders under the root directory of the project.
## Usage
To train new model
```bash
. train.sh [hs|django]
```
To use trained model for decoding test sets
```bash
. run_trained_model.sh [hs|django]
```
## Dependencies
* Theano
* vprof
* NLTK 3.2.1
* astor 0.6
## Reference
```
@inproceedings{yin17acl,
title = {A Syntactic Neural Model for General-Purpose Code Generation},
author = {Pengcheng Yin and Graham Neubig},
booktitle = {The 55th Annual Meeting of the Association for Computational Linguistics (ACL)},
address = {Vancouver, Canada},
month = {July},
url = {https://arxiv.org/abs/1704.01696},
year = {2017}
}
```