Command-Line Interface (CLI)

The cross-sensor-cal CLI provides a unified, restart-safe workflow for processing NEON flight lines from raw HDF5 to corrected and sensor-harmonized ENVI/Parquet outputs. This page documents the primary commands, flags, and usage patterns.

---

Primary command: cscal-pipeline

This is the recommended entry point for processing NEON hyperspectral data.

Example

```bash cscal-pipeline \ --base-folder output \ --site-code NIWO \ --year-month 2023-08 \ --product-code DP1.30006.001 \ --flight-lines NEON_D13_NIWO_DP1_L020-1_20230815_directional_reflectance \ --engine thread \ --max-workers 4 Required arguments FlagDescription --base-folderRoot directory for all outputs --site-codeNEON site code (e.g., NIWO, KONZ) --year-monthAcquisition month in YYYY-MM format --product-codeNEON product code (usually DP1.30006.001) --flight-linesOne or more NEON flight line identifiers Optional arguments FlagDescription --engine {thread,ray}Execution engine --max-workersNumber of parallel workers --start-atStage to begin the pipeline --end-atStage to stop the pipeline --cleanRemove intermediate files --dry-runPrint what would be run without executing Pipeline stages You may specify partial workflows: cscal-pipeline \ --start-at brdf \ --end-at convolution Valid stages: download export-envi topo brdf convolution parquet qa Engine selection Thread engine (default) Suitable for local machines and moderate-sized flight lines. Ray engine Install: pip install cross-sensor-cal[ray] Run: cscal-pipeline --engine ray ... Use Ray for: batch processing of many flight lines distributed cloud/HPC environments large-scale Parquet extraction Inspecting available options cscal-pipeline --help This prints detailed descriptions for every flag. Other CLI tools Brief mention (documented elsewhere) cscal-micasense-to-landsat debugging utilities (internal / experimental) These experimental tools may change as the package evolves. Next steps Working with Parquet outputs Pipeline stages

---