AmareshHebbar/horary_prashna-sft

Horary Astrology (Prashna) SFT Dataset

Sidereal Vedic chart cast for the moment a question is asked (not a birth moment) — same computation as Vedic natal astrology, different input semantics.

Total examples	100000
Train / Validation / Test	89916 / 5104 / 4980
Question types	4
Avg citations per example	3.942
Zero-citation examples	92

Computation

Deterministic astronomy (pyswisseph, Lahiri sidereal) — identical engine to natal Vedic astrology.

Validation status — read before trusting this at scale

Same validated engine as the Vedic natal system (see companion Jyotish-LLM project). No horary-specific validation beyond that, since the astronomy is identical.

Schema

Field	Type	Description
example_id	string	{synthetic_id}_{question_type}
synthetic_id	string	ID of the underlying synthetic input
system	string	"horary_prashna"
facts	dict	Full computed output for this system (chart/pillars/numbers depending on system)
retrieved_rules	list[dict]	Rules/observations matched for this input
question_type	string	See distribution below
user_question	string	Synthetic user question
response	string	Grounded reading — every claim traces to facts or retrieved_rules
citations	list[string]	Which rules the response draws on
source	string	"rule_matched" for all rows in this version

Question type distribution: yes_no_general: 25000, lost_object: 25000, timing: 25000, relationship_query: 25000

Usage

from datasets import load_dataset

ds = load_dataset("YOUR_USERNAME/horary_prashna-sft")
print(ds)

ex = ds["train"][0]
print(ex["user_question"])
print(ex["response"])

What this dataset does NOT claim

Like every dataset in this collection, this teaches a model to narrate a correctly-computed Horary Astrology (Prashna) reading using its own traditional rules faithfully and with citations. It does not, and cannot, contain any ground truth about whether any prediction comes true — no such data exists for any divination system. See the companion repo's top-level README for the shared "what these datasets don't claim" statement that applies across the whole collection.

Known limitations

• Responses deliberately do NOT give a yes/no answer — they describe which traditional factors are examined, since no ground truth for question outcomes exists. If your use case wants yes/no-style outputs, that requires a design decision this dataset does not make for you.
• Query-moment timestamps here are synthetic (randomly sampled 2015-2026) rather than real query moments, since no real querent data was used.
• No specialized horary techniques beyond the shared yoga/dignity rule set (e.g. no Moon void-of-course detection, no applying/separating aspect analysis).

Companion repo

Full generation pipeline (8 systems, shared scripts) — for regenerating at larger scale or extending the rule set — is the companion GitHub repo astrology-datasets (update this link once pushed).

License

Apache 2.0. All computed data is synthetic (no real people). Rule/effect text is written in-house, paraphrasing widely known traditional principles, not reproduced from any single copyrighted source.