|
TASKS_PRETTY = { |
|
"commit_message_generation": "Commit Message Generation", |
|
"bug_localization": "Bug Localization on Issue", |
|
"module_to_text": "Module-to-Text", |
|
"library_usage": "Library Usage Examples Generation", |
|
"project_code_completion": "Project-level Code Completion", |
|
"bug_localization_build_logs": "Bug Localization on Build Logs", |
|
} |
|
TASKS_PRETTY_REVERSE = {value: key for key, value in TASKS_PRETTY.items()} |
|
|
|
TASKS_DESCRIPTIONS = { |
|
"Commit Message Generation": """# Commit Message Generation\n |
|
|
|
Our Commit Message Generation benchmark π€ [JetBrains-Research/lca-cmg](https://huggingface.co/datasets/JetBrains-Research/lca-cmg) includes 163 manually curated commits from Python projects. |
|
|
|
We use the following metrics for evaluation: |
|
* [BLEU](https://huggingface.co/spaces/evaluate-metric/sacrebleu) |
|
* [ROUGE](https://huggingface.co/spaces/evaluate-metric/rouge) |
|
* [ChrF](https://huggingface.co/spaces/evaluate-metric/chrf) |
|
* [BERTScore](https://huggingface.co/spaces/evaluate-metric/berscore) |
|
|
|
For further details on the dataset and the baselines from ποΈ Long Code Arena Team, refer to `commit_message_generation` folder in [our baselines repository](https://github.com/JetBrains-Research/lca-baselines) or to our preprint (TODO). |
|
""", |
|
"Bug Localization on Issue": "cool description for Bug Localization on Issue task", |
|
"Module-to-Text": "cool description for Module-to-Text task", |
|
"Library Usage Examples Generation": "cool description for Library Usage Examples Generation task", |
|
"Project-level Code Completion": "cool description for Project-level Code Completion task", |
|
"Bug Localization on Build Logs": "cool description for Bug Localization on Build Logs task", |
|
} |
|
|