Code Generation On Django
평가 지표
Accuracy
BLEU Score
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Accuracy | BLEU Score |
|---|---|---|
| leveraging-pre-trained-language-models-for-3 | 78.50 | 89.34 |
| reranking-for-neural-semantic-parsing | 80.2 | - |
| mariancg-a-code-generation-transformer-model | 81.83 | 90.41 |
| latent-predictor-networks-for-code-generation | 62.3 | 77.6 |
| leveraging-pre-trained-language-models-for-3 | 77.95 | 88.91 |
| tranx-a-transition-based-neural-abstract | 73.7 | - |
| leveraging-pre-trained-language-models-for-3 | 76.68 | 56.55 |
| semantic-parsing-with-less-prior-and-more | 81.03 | - |
| the-impact-of-lexical-and-grammatical-1 | 81.03 | 79.86 |
| latent-predictor-networks-for-code-generation | 31.5 | 47.6 |
| leveraging-pre-trained-language-models-for-3 | 65.32 | 53.02 |