mirror of
https://github.com/acamarata/pray-calc-ml.git
synced 2026-07-01 19:34:26 +00:00
bbe1bf5cbc
Weighted least-squares calibration of Islamic prayer time depression angles from observed mosque announcement data. Uses golden-section search to minimize the sum of squared residuals independently for Fajr and Isha. Internal Jean Meeus solar ephemeris — zero runtime dependencies. API: calibrateAngles, scoreAngles, predictFajr, predictIsha. Full TypeScript, dual CJS/ESM via tsup. 32 ESM tests, 6 CJS tests, all passing on Node 20/22/24.
17 lines
889 B
Markdown
17 lines
889 B
Markdown
# Changelog
|
|
|
|
All notable changes to this project are documented here.
|
|
Format follows [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
|
|
|
|
## [1.0.0] - 2026-02-25
|
|
|
|
### Added
|
|
|
|
- `calibrateAngles()` — fit optimal Fajr/Isha depression angles to observed mosque announcement data via weighted least-squares (golden-section search)
|
|
- `scoreAngles()` — evaluate fixed depression angles against observations, returning RMS error and signed bias per prayer
|
|
- `predictFajr()` / `predictIsha()` — predict prayer times from a depression angle using an internal Jean Meeus solar ephemeris (no pray-calc dependency at runtime)
|
|
- Full TypeScript source with strict mode and dual CJS/ESM build via tsup
|
|
- `CalibrationOptions` for solver bounds and convergence control
|
|
- Graceful handling of Fajr-only or Isha-only datasets
|
|
- 32-test ESM suite and 6-test CJS suite
|
|
- CI matrix: Node 20, 22, 24
|