Bachelor thesis code (GIU, Winter 2026). Project brief, rules and model roster: CLAUDE.md.
| File | What |
|---|---|
| docs/progress.md | Run notes: what each run did, what changed, what it showed |
| docs/decisions.md | Decisions log (D-NNN) |
| docs/data.md | Data sources and versions (D1–D4) |
| docs/tables/split_summary.md | Classifier scores per run: validation, dev, test, per dialect; leaderboard submissions |
| docs/tables/quality_summary.md | Quality measurements of data sets (Q) |
| docs/tables/scorer_summary.md | Scorer validation against human gold (S) |
| docs/tables/generator_summary.md | Generator feasibility checks (G) |
| thesis/ | Report notes |
Tables in docs/tables/ are generated from results/ and never edited by hand:
python -m src.evaluation.summarize_runs # split_summary
python -m src.quality.summarize_quality # quality, scorer and generator summaries
Run IDs and result folders are registered in configs/runs.json.
Bachelor thesis code (GIU, Winter 2026). Project brief, rules and model roster: CLAUDE.md.
| File | What |
|---|---|
| docs/progress.md | Run notes: what each run did, what changed, what it showed |
| docs/decisions.md | Decisions log (D-NNN) |
| docs/data.md | Data sources and versions (D1–D4) |
| docs/tables/split_summary.md | Classifier scores per run: validation, dev, test, per dialect; leaderboard submissions |
| docs/tables/quality_summary.md | Quality measurements of data sets (Q) |
| docs/tables/scorer_summary.md | Scorer validation against human gold (S) |
| docs/tables/generator_summary.md | Generator feasibility checks (G) |
| thesis/ | Report notes |
Tables in docs/tables/ are generated from results/ and never edited by hand:
python -m src.evaluation.summarize_runs # split_summary
python -m src.quality.summarize_quality # quality, scorer and generator summaries
Run IDs and result folders are registered in configs/runs.json.