Leveraging Reviewer Experience in Code Review Comment Generation

(Replication Package)

This is the replication package for journal paper "Leveraging Reviewer Experience in Code Review Comment Generation".

Authors

1. Hong Yi Lin (University of Melbourne)
2. Patanamon Thongtanunam (University of Melbourne)
3. Christoph Treude (Singapore Management University)
4. Michael Godfrey (University of Waterloo)
5. Chunhua Liu (University of Melbourne)
6. Wachiraphan Charoenwet (University of Melbourne)

List of Contents

• Manual_Evaluation 
  • Accuracy annotations
  • Informativeness annotations
  • Comment category annotations
  • Manual annotation guidelines
  • 100 random samples
• ELF_AVG 
  •  Model checkpoints for ELF_AVG strategy (Repository, Subsystem, Package)
• ELF_ACO 
  • Model checkpoints for ELF_ACO strategy (Repository, Subsystem, Package)
• ELF_MAX 
  • Model checkpoints for ELF_MAX strategy (Repository, Subsystem, Package)
• ELF_RSO
  • Model checkpoints for ELF_RSO strategy (Repository, Subsystem, Package)
• Oversampling
  • Model checkpoints for Experience-Aware Oversampling (Repository)
• CodeReviewer
  • Model checkpoints for the original CodeReviewer (Fine-tuned on our dataset)
• Predictions
  • All model generated predictions for test set
• Repository_History
  • Pull request and commit histories for all repositories in training, validation and test set
• Code_Review_Dataset_Tagged
  • Cleaned training, validation and test set including tagged ownership ratios
• Top10_B4_Delta
  • Top 10 generations for each ELF model vs CodeReviewer in terms of BLEU-4
• ELF_Code
  • Fine-tuning and testing scripts for ELF (Adapted from CodeReviewer)
• Preliminary_Analysis
  • Annotations of real code reviews provided by both experienced and inexperienced reviewers in the training set.
  • 89 samples for each reviewer type