spectral-library 0.1.0

Highlights

• First public release of the spectral-library package and CLI
• Prepared-runtime build, validation, and retrieval-based mapping workflows
• Stable mapping output modes for target-sensor, VNIR, SWIR, and full-spectrum outputs
• Batch mapping support in both the Python API and CLI
• Regression-baseline benchmarking and prepared-runtime contract validation

Public Surface

• Python API: prepare_mapping_library, validate_prepared_library, SpectralMapper, MappingResult, BatchMappingResult, PreparedLibraryManifest, SensorSRFSchema, benchmark_mapping
• CLI: prepare-mapping-library, validate-prepared-library, map-reflectance, map-reflectance-batch, benchmark-mapping

Documentation

• Published docs site: https://marcyin.github.io/spectral_library/
• Quickstart: docs/mapping_quickstart.md
• Mathematical foundations: docs/theory.md
• Official sensor examples: docs/official_sensor_examples.md
• Internal pipeline reference: docs/internal_build_pipeline.md
• Production release standard: docs/production_release_standard_plan.md

Compatibility Notes

• Prepared runtimes use schema version 1.0.0
• Package license: MIT for repository software; third-party datasets and derived artifacts retain their upstream terms
• Public output modes are: target_sensor, vnir_spectrum, swir_spectrum, full_spectrum
• Python support in CI: 3.9, 3.10, 3.11, 3.12