expand_fits#

mufasa.master_fitter.expand_fits(reg, ncomp, lnk_thresh=5, max_iter=5, r_expand=1, fill_mask=None, lnkmap=None, multicore=True, snr_min=0, save_para=True, return_niter=False)[source]#

Expand fits in a region by incrementally fitting pixels beyond a defined log-likelihood threshold boundary.

This function iteratively expands fitted regions by attempting to fit pixels beyond a boundary set by the relative log-likelihood threshold.

Parameters:

  • reg (Region object) – The region object containing the data cube and the fitting model.

  • ncomp (int) – Number of components in the model to fit in each pixel.

  • lnk_thresh (float, optional) – Relative log-likelihood threshold used to define the boundary for expansion. Pixels with log-likelihood values above this threshold are considered for expansion (default is 5).

  • max_iter (int, optional) – Maximum number of expansion iterations to perform. Each iteration attempts to fit the next layer of pixels beyond the current boundary (default is 5).

  • r_expand (int, optional) – Number of pixels to expand in each direction during each iteration (default is 1).

  • fill_mask (ndarray, optional) – Boolean array that defines the outer boundary for expansion. Expansion is limited to pixels within this mask (default is None, which implies no additional mask constraint).

  • lnkmap (ndarray, optional) – A precomputed 2D array (i.e., map) of relative log-likelihood values. Supplying this map saves computation time by avoiding recalculating these values. If None, the function will compute the map as needed (default is None).

  • multicore (bool or int, optional) – Whether to use multiple cores for parallel processing. If True, automatically selects available cores; if an integer is provided, it specifies the number of cores (default is True).

  • snr_min (float, optional) – Minimum signal-to-noise ratio threshold for successful fits. Fits below this threshold are not considered “good” (default is 0).

  • save_para (bool, optional) – If True, saves updated parameter maps after fitting. Useful for keeping track of intermediate fit results (default is True).

  • return_niter (bool, optional) – If True, returns both the total number of successfully fitted pixels and the number of iterations performed (default is False).

Returns:

If return_niter is False, returns the total number of successfully fitted pixels across all iterations. If return_niter is True, returns a tuple containing the total number of successfully fitted pixels and the number of iterations completed.

Return type:

int or tuple of (int, int)

Examples

>>> n_good_total = expand_fits(reg, ncomp=2, lnk_thresh=10, max_iter=3, r_expand=1, multicore=4)
>>> print(f"Total good fits: {n_good_total}")
On this page

This Page