gaussian_random_field.grf_module_entrypoint

Entrypoint for GRF module.

Functions

main(mesh_name, data_path, output_path[, ...])

Run the Gaussian Random Field model.

main(mesh_name: str, data_path: ZarrType, output_path: Path, prior_sigma_1: float = 1.0, prior_sigma_2: float = 0.01, grid_res_x: int = 100, grid_res_y: int = 100) → None

Run the Gaussian Random Field model.

Parameters:
  • mesh_name (str) – The name of the mesh to load and use.

  • data_path (ZarrType) – Path of the input GeoZarr file.

  • output_path (Path) – Path of the output GeoZarr file.

  • prior_sigma_1 (float, optional) – Parameter characterizing the expected error in the data.

  • prior_sigma_2 (float, optional) – Probability that error exceeds prior_sigma_1.

  • grid_res_x (int, optional) – Number of grid points in the x direction.

  • grid_res_y (int, optional) – Number of grid points in the y direction.

Raises:

ConstantFieldError – If no variable in any timestep has any spatial variation. A variable with none is filled with the constant it already holds, and any other variable at that timestep is still fitted; this is raised only when nothing in the record was fitted at all.