CoolFace
Datasetpublic

CodeIsAbstract/sanskrit-samas-v1

Sanskrit Samas (Compound) Dataset — V1 Grammar-grounded training data for Sanskrit samas (compounds) covering all 7 samasa types, with laukik vigraha (natural paraphrase) and alaukik vigraha (Pāṇinian sUP analysis) on every row. Companion to sanskrit-sandhi-boundaries-v2 (sentence-level external sandhi). Merge both for a full sandhi+samas boundary training set. Dataset Rows: 258,408 Checksum: e7c0a804c109 Builder: benchmarks/build_samas_data.py (deterministic… See the full description on the dataset page: https://huggingface.co/datasets/CodeIsAbstract/sanskrit-samas-v1.

sourceHugging Facemitupdated 29d agoView on Hugging Face
0likes47downloads
Dataset Card

Sanskrit Samas (Compound) Dataset — V1

Grammar-grounded training data for Sanskrit samas (compounds) covering all 7 samasa types, with laukik vigraha (natural paraphrase) and alaukik vigraha (Pāṇinian sUP analysis) on every row.

Companion to `sanskrit-sandhi-boundaries-v2` (sentence-level external sandhi). Merge both for a full sandhi+samas boundary training set.

Dataset

  • —Rows: 258,408
  • —Checksum: e7c0a804c109
  • —Builder: benchmarks/build_samas_data.py (deterministic, seed 42)
  • —Format: JSONL (one compound per line)

Schema (per row)

FieldTypeDescription
surfacestringNatural merged SLP1 surface (internal sandhi applied at the seam)
devanagaristringDevanagari rendering
pieceslist[str]Surface partition at the compound seam (splitter contract; "".join == surface)
bndlist[int]Boundary positions (training format; last-char indices)
gold_wordslist[str]Laukik vigraha — natural paraphrase with explicit case forms
gold_alaukikstringAlaukik vigraha — Pāṇinian analysis: stems + sUP tags (e.g. rAma + Nas + puruSa + su~)
samasa_typestringOne of the 7 types
stem1, stem2stringThe constituent stems

Samasa types covered

TypeMeaningExample (laukik)n
tatpuruṣadeterminative (X of Y)rājasya puruṣaḥ62,893
karmadhārayaappositional (X that is Y)nīlam utpalam62,893
dvandvacopulative (X and Y)rāmaḥ ca kṛṣṇaḥ ca59,924
bahuvrīhiexocentric "having X"cakraṃ pāṇau yasya saḥ60,172
avyayībhāvaadverbialśaktim anatikramya7,097
amreḍitareduplicativedine dine5,091
dvigunumericaltrayo lokāḥ338

Construction (Pāṇinian)

  1. 1.First member(s) lose their case endings → bare stem (prātipadika).
  2. 2.Last member keeps its own case ending.
  3. 3.Internal sandhi applies at the seam (same rules as external sandhi).
  4. 4.The compound takes ONE case ending on the last member.

Alaukik vigraha is generated from vidyut prakriya derivation history — the exact sUP tags (su~, am, wA, Ne, Nasi~, Nas, Ni × Eka/Dvi/Bahu) are recovered from the Pāṇinian derivation steps.

Integrity

  • —0 rows fail "".join(pieces) == surface.
  • —Every seam is rejoin-verified by the sandhi engine.
  • —Builder deterministic (seed 42): reproducible byte-for-byte.