smlmlp.globloc_fit module

smlmlp.globloc_fit(crops, X0, Y0, /, channels_models, channels_fit_inits, *, optimizer='lm', estimator='mle', distribution='poisson', channels_pixels_nm=1.0, channels_gains=1.0, channels_QE=1.0, cuda=False, parallel=False)[source]

Fit global localizations from channel crops using joint fitting.

Uses funclp.JointFunction to fit all crops simultaneously, sharing position parameters across channels.

Parameters:

  • crops (sequence of ndarray) – Crop stacks to fit, one per channel, shaped (N, Y, X).

  • X0 (sequence of ndarray) – Crop x origins in pixels.

  • Y0 (sequence of ndarray) – Crop y origins in pixels.

  • channels_models (sequence of str) – Model per channel, one of "gauss", "isogauss", "spline".

  • channels_fit_inits (sequence of dict) – Initial fit parameters per channel.

  • optimizer (str, optional) – Optimizer key.

  • estimator (str, optional) – Estimator key.

  • distribution (str, optional) – Distribution key used by the estimator.

  • channels_pixels_nm (float or sequence, optional) – Pixel size specification per channel.

  • channels_gains (float or sequence, optional) – Gain value(s) used for fitted amplitudes.

  • channels_QE (float or sequence, optional) – Quantum efficiency value(s) used for fitted amplitudes.

  • cuda (bool, optional) – Whether to use CUDA execution.

  • parallel (bool, optional) – Whether to use parallel execution.

Returns:

A tuple (mux, muy, info) where:

  • mux is the concatenated x localization array in nanometers,

  • muy is the concatenated y localization array in nanometers,

  • info is a dictionary with fitted parameter arrays.

Return type:

tuple

Examples

>>> import numpy as np
>>> crops = [np.random.rand(2, 7, 7).astype(np.float32)]
>>> x0 = [np.array([10, 20], dtype=np.float32)]
>>> y0 = [np.array([30, 40], dtype=np.float32)]
>>> models = ["gauss"]
>>> inits = [{"sigx": 90.0, "sigy": 90.0, "theta": 0.0, "theta_fit": False}]
>>> mux, muy, info = globloc_fit(
...     crops, x0, y0,
...     channels_models=models,
...     channels_fit_inits=inits,
...     channels_pixels_nm=[(100.0, 100.0)],
... )
>>> mux.shape == muy.shape
True

>>> crops = [np.random.rand(2, 7, 7).astype(np.float32)]
>>> x0 = [np.array([10, 20], dtype=np.float32)]
>>> y0 = [np.array([30, 40], dtype=np.float32)]
>>> models = ["isogauss"]
>>> inits = [{"sig": 90.0}]
>>> mux, muy, info = globloc_fit(
...     crops, x0, y0,
...     channels_models=models,
...     channels_fit_inits=inits,
...     channels_pixels_nm=[(100.0, 100.0)],
... )
>>> info['sigma'].ndim
1