Skip to content

Module Reference Map

This page is not a full API reference. Instead, it is a guide to where the most important repository concepts currently live.

Core pado Modules

  • pado.light: complex light-field representation
  • pado.optical_element: optical elements and modulation
  • pado.propagator: free-space propagation methods
  • pado.material: material definitions
  • pado.math: numerical helpers

Compatibility Bridge

  • pado.display: LUT-based LCOS/SLM encoding bridge kept in the pado namespace for compatibility while higher-level SLM workflows move toward pado_hologram

pado_hologram Modules

  • pado_hologram.config: source and propagation specifications
  • pado_hologram.backends: optional custom-kernel backend helpers, including NVIDIA Warp
  • pado_hologram.targets: target intensity representations
  • pado_hologram.losses: reconstruction losses and metrics
  • pado_hologram.slm: framework-level LCOS/SLM helpers
  • pado_hologram.pipeline: single-plane and multi-plane workflow composition
  • pado_hologram.algorithms: GS and DPAC helpers
  • pado_hologram.primitive_based: gsplat-free primitive-scene Gaussian baselines, depth-aware and alpha-wave-blending variants, exact GWS/RPWS paths, backend selection, and preset/JSON scene ingestion
  • pado_hologram.devices: SLM support and optional camera/observation transforms
  • pado_hologram.experiments: experiment execution utilities and config composition
  • pado_hologram.experiment: backward-compatible re-export for older experiment imports
  • pado_hologram.hydra_app: Hydra CLI entry point

New Internal Layout

The package is also beginning to migrate toward a more scalable internal split:

  • pado_hologram.core: specs, targets, losses, and pipelines
  • pado_hologram.devices: SLM-facing abstractions plus optional camera observation helpers
  • pado_hologram.phase_only: phase-only CGH algorithms such as GS and DPAC
  • pado_hologram.primitive_based: primitive-based renderer baselines, preset/JSON scene ingestion, the first vectorized splat path, a depth-aware gaussian_wave baseline, an hsplat-inspired gaussian_wave_awb path for alpha wave blending, and an exact GWS-backed RPWS baseline with structured random phase and time-multiplexed intensity averaging, all without an external gsplat dependency
  • pado_hologram.experiments: experiment runners and Hydra-facing wrappers
  • pado_hologram.representations: primitive-based scene representations
  • pado_hologram.neural: neural holography, calibration, and capture-oriented scaffolds

These newer paths are meant to make future additions such as primitive-based CGH or neural holography easier to place without overloading the top-level package.

Where to Look Next

If you need exhaustive API details, use the Sphinx documentation that remains in docs/source.

This MkDocs page is meant to answer the practical question: "which module should I open first?"