FITS_tools.cube_regrid.spectral_smooth_cube(cube, kernelwidth, kernel=<class 'astropy.convolution.kernels.Gaussian1DKernel'>, cubedim=0, parallel=True, numcores=None, use_fft=False, **kwargs)[source] [edit on github]

Parallelized spectral smoothing


cube : ndarray

A data cube, with ndim=3

kernelwidth : float

Width of the kernel. Defaults to Gaussian.

kernel : Kernel1D

A 1D kernel from astropy

cubedim : int

The axis NOT to map across, i.e. the spectral axis. If you have a normal FITS data cube with AXIS1=RA, AXIS2=Dec, and AXIS3=wavelength, for example, cubedim is 0 (because axis3 -> 0, axis2 -> 1, axis1 -> 2)

numcores : int

Number of cores to use in parallel-processing.

use_fft : bool

Use convolve_fft or convolve?

kwargs : dict

Passed to astropy.convolution.convolve