pyfli.solver.binned_fitter#

Bin image cubes spatially and fit the binned FLI data.

This module belongs to pyfli.solver and is part of PyFLI least-squares, maximum- likelihood, CPU, GPU, binned, and global FLI fitting routines. Public API includes classes FLIBinner and BinnedFLIFitter.

Classes

BinnedFLIFitter(processor_instance[, bin_radius])

Fit spatially binned FLI data with an existing processor.

FLIBinner([bin_radius])

Apply spatial binning to FLI image and IRF cubes.
class FLIBinner(bin_radius=1)[source]#

Bases: object

Apply spatial binning to FLI image and IRF cubes. It reduces noise by aggregating neighboring pixels before fitting.

Parameters:

bin_radius (int) – Radius of the spatial binning neighborhood in pixels.

apply_binning(image_cube, irf_cube)[source]#

Performs spatial binning using constant padding to maintain original image dimensions.

Parameters:
Return type:

tuple[Any, …]

get_binned_data()[source]#

Returns the binned cubes for manual inspection.

Return type:

tuple[Any, …]

class BinnedFLIFitter(processor_instance, bin_radius=1)[source]#

Bases: object

Fit spatially binned FLI data with an existing processor. It wraps binning, mask propagation, processor dispatch, and result saving for binned datasets.

Parameters:

  • processor_instance (Any) – Optional processor reused for pixel-level fitting or reconstruction.

  • bin_radius (int) – Radius of the spatial binning neighborhood in pixels.

fit(b_img, b_irf, mask=None, data_name='Binned_Dataset', **kwargs)[source]#

Unified entry point using Duck-Typing. Accepts PRE-BINNED data cubes.

Parameters:
Return type:

ndarray

save_results(dataset, folder='results')[source]#

Pass-through to the underlying processor’s optimized save logic.

Parameters:
Return type:

None