hispecdrp.primitives.spectral_extraction.extract_optimal

Optimal spectral extraction (Horne), with optional iso-λ shear.

Functions

_extract_optimal_numba(input_data, ...[, ...])

_extract_optimal_sheared_numba(input_data, ...)

_horne_accumulate(D, D_var, D_var_poisson, ...)

Shared Horne weighted sums for one spectral sample (in-place masks).

_interp_spectral_sample(arr, x_frac, y, ...)

Linear interpolate arr along the spectral axis at cross-disp y.

_nearest_dq_sample(dq, x_frac, y, ...)

extract_spectrum_optimal(input_data, ...[, ...])

Optimal spectral extraction.

_extract_optimal_numba(input_data, input_err, input_dq, input_var_rnoise, input_var_poisson, input_mid_time, trace_positions, profile, extraction_aperture, pix_range, spec_output, err_output, var_rnoise_output, var_poisson_output, dq_output, mid_time_output, orders_are_vertical=True)[source]
_extract_optimal_sheared_numba(input_data, input_err, input_dq, input_var_rnoise, input_var_poisson, input_mid_time, trace_positions, profile, shear, extraction_aperture, pix_range, spec_output, err_output, var_rnoise_output, var_poisson_output, dq_output, mid_time_output, orders_are_vertical=True)[source]
_horne_accumulate(D, D_var, D_var_poisson, D_var_rnoise, D_dq, P, T, W, M_in)[source]

Shared Horne weighted sums for one spectral sample (in-place masks).

_interp_spectral_sample(arr, x_frac, y, orders_are_vertical)[source]

Linear interpolate arr along the spectral axis at cross-disp y.

_nearest_dq_sample(dq, x_frac, y, orders_are_vertical)[source]
extract_spectrum_optimal(input_data, input_err, input_dq, input_var_rnoise, input_var_poisson, input_mid_time, trace_positions, profile, extraction_aperture, pix_range=None, spec_output=None, err_output=None, var_rnoise_output=None, var_poisson_output=None, dq_output=None, mid_time_output=None, orders_are_vertical=True, shear=None)[source]

Optimal spectral extraction.

Parameters:

shear (np.ndarray | None) – Optional iso-λ shear ∂(spectral)/∂(cross-disp) per spectral sample (same length as trace_positions). When set, data/profile are sampled along x' = x + shear[x] * (y - y_c) before the Horne sum.

get_extraction_window(yc, window_size, n_pixels=-1)[source]

Computes the fractional overlap of pixels with the extraction window.

Parameters:
  • yc (float) – Center of the extraction window (in pixel coordinates).

  • window_size (float) – Size of the extraction window (in pixels) centered in yc.

  • n_pixels (int, optional) – When >= 0, clip candidate indices to [0, n_pixels - 1] so callers never receive out-of-bounds cross-dispersion indices (avoids Numba OOB segfaults when trace positions are extrapolated or non-finite).

Returns:
  • idx (np.ndarray) – Indices of the pixels that might have overlap with the extraction window.

  • W (np.ndarray) – Fractional overlap of each pixel in idx with the extraction window.