apcl/jam_contextsum

## Jam-Contextsum
Jam-Contextsum dataset contains the method, the summary that explains why the method exists, caller of the method, and the summary of the call graph method generated by using [jam-cgpt](https://huggingface.co/apcl/Jam-CGPT).

## Jam-Contextsum dataset files

| Filename | Description |
| ------- | ------- |
|contextsum.pkl|methods with the summary that explains why the method exists, caller of the method, and the summary of the caller generated by using [jam-cgpt](https://huggingface.co/apcl/Jam-CGPT) |
|unique_methods.pkl | all of the unique methods|

Note that we tokenize our data and extract the summary on why the method exists using scripts provided in our [github repository](https://github.com/apcl-research/jam-contextsum).

## Human study results
- All results are in ``human_study_result.tar.gz``.
- The name of the directory shows the type of comparison e.g. "gemini_context_vs_gemini_baseline" will be gemini context vs. gemini baseline in the paper.
- Each directory has four files except rationale -- functions.csv, summary.csv, Q1_survey.csv, and Q2_survey.csv. Q1_survey.csv has the data from Q1 in the survey and Q2_survey.csv has the data from Q2 in the survey (Functions for the study should be the same in each directory).
- In the rationale directory, rank_results.csv is the file for the ranking results and the rationale_results.csv is the reseults for rationale. It also has function.csv and summary.csv for functions and summary respectively.