The Ship Design Decision Dataset (SDD Dataset) contains cleaned Chinese natural-language requests paired with typed decisions for constrained ship-form editing. SDDBench is the fixed 5,000-record evaluation subset used in the NL2Hull paper.
Dataset Card
The Ship Design Decision Dataset (SDD Dataset) contains cleaned Chinese natural-language requests paired with typed decisions for constrained ship-form editing. SDDBench is the fixed 5,000-record evaluation subset used in the NL2Hull paper. - train.jsonl: 88,604 records - validation.jsonl: 22,758 records - test.jsonl: 23,196 records - test5k.jsonl: frozen 5,000-record benchmark subset - languagesamples.jsonl: source language snapshot The split policy separates hull families across training, validation, and test sets. The training split uses DTC, DTMB 5415, KCS, KVLCC2, containership, frigate, and NPL variants; validation uses S-175 and Series 60; test uses Wigley and workboat hulls. Each…
Excerpt from the card by Wenhua Huo.
Details
- Repository
- wenhuahuo/ship-design-decisions
- Publisher
- Wenhua Huo
- Task category
- Text generation
- Tags
- ship-design, natural-language, typed-decision
- Size category
- 100K<n<1M
- Languages
- zh
- Revision
- 3f952a5a740fd8067caf25b4f9ce4a236fa77fa4
- Last updated
- 2026-10-09
Files
12 files, 510.8 MB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| language_samples.jsonl | Data | 57.1 MB | ad871d0164de |
| manifest.json | Data | 1.6 KB | — |
| selection_manifest.json | Data | 1.1 KB | — |
| source_cleaning_manifest.json | Data | 1.2 KB | — |
| test.jsonl | Data | 74.5 MB | 88f2797cf8a9 |
| test_5k.jsonl | Data | 16.1 MB | 6e340ea956e2 |
| test_5k_manifest.json | Data | 13.1 KB | — |
| train.jsonl | Data | 289.7 MB | faa88587f803 |
| validation.jsonl | Data | 73.4 MB | e5b59821c53f |
| DATASET_README.md | Documentation | 582 B | — |
| README.md | Documentation | 2.9 KB | — |
| .gitattributes | Repository | 2.8 KB | — |
License and Download
- License
- Not stated by the source
- Access
- No access gate
Released by Wenhua Huo through its official repository on Hugging Face.