Add Module-to-Text description

#1
by jdev8 - opened
Files changed (1) hide show
  1. src/tasks_content.py +12 -1
src/tasks_content.py CHANGED
@@ -26,7 +26,18 @@ TASKS_DESCRIPTIONS = {
26
  **Note.** The leaderboard is sorted by ROUGE-1 metric by default.
27
  """,
28
  "bug_localization": "cool description for Bug Localization on Issue task",
29
- "module_to_text": "cool description for Module-to-Text task",
 
 
 
 
 
 
 
 
 
 
 
30
  "library_usage": "cool description for Library Usage Examples Generation task",
31
  "project_code_completion": "cool description for Project-level Code Completion task",
32
  "bug_localization_build_logs": "cool description for Bug Localization on Build Logs task",
 
26
  **Note.** The leaderboard is sorted by ROUGE-1 metric by default.
27
  """,
28
  "bug_localization": "cool description for Bug Localization on Issue task",
29
+ "module_to_text": """# Module-to-Text\n
30
+
31
+ Our Module-to-Text benchmark ๐Ÿค— [JetBrains-Research/lca-module-to-text](https://huggingface.co/datasets/JetBrains-Research/lca-module-to-text) includes 206 manually curated text files describing modules from different Python projects.
32
+
33
+ We use the following metrics for evaluation:
34
+ * [ROUGE](https://huggingface.co/spaces/evaluate-metric/rouge)
35
+ * [ChrF](https://huggingface.co/spaces/evaluate-metric/chrf)
36
+ * [BERTScore](https://huggingface.co/spaces/evaluate-metric/bertscore)
37
+ * ChatGPT-Turing-Test
38
+
39
+ For further details on the dataset and the baselines from ๐ŸŸ๏ธ Long Code Arena Team, refer to `module2text` folder in [our baselines repository](https://github.com/JetBrains-Research/lca-baselines) or to our preprint (TODO).
40
+ """,
41
  "library_usage": "cool description for Library Usage Examples Generation task",
42
  "project_code_completion": "cool description for Project-level Code Completion task",
43
  "bug_localization_build_logs": "cool description for Bug Localization on Build Logs task",