Revisiting the Impact of Pursuing Modularity for Code Generation

Official Repository for "Revisiting the Impact of Pursuing Modularity for Code Generation" [Paper(arXiv)]

Deokyeong Kang, Kijung Seo, Taeuk Kim. Accepted to EMNLP 2024 Findings

Contents

• apps: source codes for APPS dataset
  • In-Context Learning: icl.py
• codecontests: source codes for CodeContests dataset
  • Modularity score (MoS) metric: utils/utils.py
  • In-Context Learning: icl.py
  • Fine-tuning: ft.py
  • Correlation experiment: icl_corr.py
  • Perplexity experiment: ppl.py