Skip to content

spectrocube

SpectroCube is a lightweight Python standard for calibrated spectroscopic datasets. It defines a single class, SpectroCube, that wraps an xarray Dataset with required metadata, a validated data structure, and NetCDF serialization.

The goal is a portable, self-describing file format: instrument packages produce SpectroCube files; analysis packages consume them — without sharing any instrument-specific code.

Package scope

This package contains:

  • The SpectroCube class and its constructors
  • The ValidationReport class
  • Save / load via NetCDF

This package intentionally does not contain:

  • Fitting or line identification
  • Plotting or visualization
  • Instrument-specific extraction or calibration logic
  • Analysis workflows

Data model at a glance

A SpectroCube always has:

  • An intensity data variable (any shape, must include wavelength as a dimension)
  • A wavelength coordinate (1D, nm, monotonically increasing)
  • Five required global attributes: spectrocube_version, instrument_id, calibration_type, intensity_units, wavelength_medium

The underlying xr.Dataset is always accessible via .ds for direct xarray operations.