pycsa.core

The src subpackage contains the mathematical modules and their accompanying utilities for the constrained spectral approximation method.

Modules

buffer_pool	Dynamic buffer pool for reusing NumPy arrays across multiple computations.
delaunay	Delaunay decomposition and land-cell selection.
fourier
hyperparams	Hyperparameter selection for the structured Tikhonov prior.
io	Input/Output routines
lin_reg	Linear regression module with buffer pool and sparse solver support
mode_selection	Pluggable mode selectors for the FA -> SA bridge.
physics
priors	Pluggable Tikhonov priors for the linear-fit step.
reconstruction	Reassembly of masked reconstruction vectors back to 2D fields.
tile_cache	Topography tile caching system for efficient parallel processing.
utils	This module contains miscellaneous helper functions and classes
validation	Held-out validation utilities for the structured prior.
var	Deprecation shim for the old monolithic var.py.