CoCoNut-Artifact

KO | EN

gitlite — search

#javascript #php #java #c-plus-plus #game #linux #swift #python #ios #shell #r #ansible

CoCoNut-Artifact

★ 63 Open GitHub ↗

For our ISSTA20 paper "CoCoNuT: Combining Context-Aware Neural Translation Models using Ensemble for Program Repair" by Thibaud Lutellier, Hung Viet Pham, Lawrence Pang, Yitong Li, Moshi Wei and Lin Tan

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# CoCoNut-Artifact
Combining Context-Aware Neural Translation Models using Ensemble for Program Repair. Artefact repo

If you use CoCoNuT for academic purposes, please cite the following publication:
```
@article{lutellier2020coconut,
  title={CoCoNuT: Combining Context-Aware Neural Translation Models using Ensemble for Program Repair},
  author={Lutellier, Thibaud and Pham, Viet Hung and Pang, Lawrence and Li, Yitong and Wei, Moshi and Tan, Lin},
  booktitle={Proceedings of the 28th ACM SIGSOFT International Symposium on Software Testing and Analysis},  
  year={2020}
}
```

Results directory contains the correct patches generated by CoCoNuT.

fairseq-context directory contains an updated fairseq API that also implements the Context-aware architecture.

To use it, you can follow https://fairseq.readthedocs.io/en/latest/models.html#module-fairseq.models.fconv
To use the new Context-aware model, you need to add the argument "--use-context" and choose the "fconv_context" for the architecture.

For example:
```
python $fairseq_dir/train.py --use-context --fp16 --save-dir $trg_dir/model --arch fconv_context --distributed-world-size 1 --encoder-embed-dim 250 --decoder-embed-dim 250 --decoder-out-embed-dim 250 --encoder-layers '[(256,3)] * 7' --decoder-layers '[(256,3)] * 7' --dropout 0.2 --clip-norm 0.1 --lr 0.25 --min-lr 1e-4 --momentum 0.99 --max-epoch 1 --batch-size 32 $trg_dir/bin
```

CoCoNut-Artifact

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README