nfft无法与奇数图像尺寸一起使用！在ehtim

Question

我正在尝试使用Python软件包EHTIM获取合成VLBI图像（使用了事件地平线望远镜的数据）。示例符合文件，一切正常。但是，当我尝试添加一个黑洞时，借助射线追踪（使用WCS坐标网格）生成的图像时，

from __future__ import division
from __future__ import print_function
import matplotlib.pyplot as plt
import numpy as np
import ehtim as eh
from   ehtim.calibrating import self_cal as sc
import time
#from  ehtim.plotting import self_cal as sc
plt.close('all')

ttype = 'direct'

# Load the image and the array
im = eh.image.load_fits('disk_SchildWH.fits')
eht = eh.array.load_txt('EHT2017.txt')

# Look at the image
im.display()

# Observe the image
# tint_sec is the integration time in seconds, and tadv_sec is the advance time between scans
# tstart_hr is the GMST time of the start of the observation and tstop_hr is the GMST time of the end
# bw_hz is the  bandwidth in Hz
# sgrscat=True blurs the visibilities with the Sgr A* scattering kernel for the appropriate image frequency
# ampcal and phasecal determine if gain variations and phase errors are included
tint_sec = 5
tadv_sec = 600
tstart_hr = 0
tstop_hr = 24
bw_hz = 4e9
obs = im.observe(eht, tint_sec, tadv_sec, tstart_hr, tstop_hr, bw_hz,
                 sgrscat=False, ampcal=True, phasecal=False)

npix = 32
fov = 1.5*im.xdim * im.psize # slightly enlarge the field of view


# Resolution
beamparams = obs.fit_beam() # fitted beam parameters (fwhm_maj, fwhm_min, theta) in radians
res = obs.res() # nominal array resolution, 1/longest baseline
print("Clean beam parameters: " , beamparams)
print("Nominal Resolution: " ,res)

# Export the visibility data to uvfits/text
#obs.save_txt('obs.txt') # exports a text file with the visibilities
#obs.save_uvfits('obs.uvp') # exports a UVFITS file modeled on template.UVP

# Generate an image prior
npix = 128
fov = 1*im.fovx()
zbl = im.total_flux() # total flux
prior_fwhm = 200*eh.RADPERUAS # Gaussian size in microarcssec
emptyprior = eh.image.make_square(obs, npix, fov)
flatprior = emptyprior.add_flat(zbl)
gaussprior = emptyprior.add_gauss(zbl, (prior_fwhm, prior_fwhm, 0, 0, 0))

# Image total flux with bispectrum
flux = zbl
tt = time.time()
out  = eh.imager_func(obs, gaussprior, gaussprior, flux,
                      d1='bs', s1='simple',
                      alpha_s1=1, alpha_d1=100,
                      alpha_flux=100, alpha_cm=50,
                      maxit=100, ttype=ttype, show_updates=False)

# Blur the image with a circular beam and image again to help convergance
out = out.blur_circ(res)
out = eh.imager_func(obs, out, out, flux,
                d1='bs', s1='tv',
                alpha_s1=1, alpha_d1=50,
                alpha_flux=100, alpha_cm=50,
                maxit=100,ttype=ttype, show_updates=False)

out = out.blur_circ(res/2.0)
out = eh.imager_func(obs, out, out, flux,
                d1='bs', s1='tv',
                alpha_s1=1, alpha_d1=10,
                alpha_flux=100, alpha_cm=50,
                maxit=100,ttype=ttype, show_updates=False)

print ("total time: ", time.time() - tt)

outblur = out.blur_gauss(beamparams, 0.5)
out.display()

我遇到了一个错误，

Generating empty observation file . . . 
Producing clean visibilities from image with nfft FT . . . 

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
/tmp/ipykernel_196400/2864228422.py in <module>
     32 tstop_hr = 24
     33 bw_hz = 4e9
---> 34 obs = im.observe(eht, tint_sec, tadv_sec, tstart_hr, tstop_hr, bw_hz,
     35                  sgrscat=False, ampcal=True, phasecal=False)
     36 

~/.local/lib/python3.10/site-packages/ehtim/image.py in observe(self, array, tint, tadv, tstart, tstop, bw, mjd, timetype, polrep_obs, elevmin, elevmax, ttype, fft_pad_factor, fix_theta_GMST, sgrscat, add_th_noise, jones, inv_jones, opacitycal, ampcal, phasecal, frcal, dcal, rlgaincal, stabilize_scan_phase, stabilize_scan_amp, neggains, tau, taup, gain_offset, gainp, dterm_offset, caltable_path, seed, sigmat, verbose)
   2497 
   2498         # Observe on the same baselines as the empty observation and add noise
-> 2499         obs = self.observe_same(obs, ttype=ttype, fft_pad_factor=fft_pad_factor,
   2500                                 sgrscat=sgrscat, add_th_noise=add_th_noise,
   2501                                 jones=jones, inv_jones=inv_jones,

~/.local/lib/python3.10/site-packages/ehtim/image.py in observe_same(self, obs_in, ttype, fft_pad_factor, sgrscat, add_th_noise, jones, inv_jones, opacitycal, ampcal, phasecal, frcal, dcal, rlgaincal, stabilize_scan_phase, stabilize_scan_amp, neggains, taup, gain_offset, gainp, dterm_offset, caltable_path, seed, sigmat, verbose)
   2353             np.random.seed(seed=seed)
   2354 
-> 2355         obs = self.observe_same_nonoise(obs_in, sgrscat=sgrscat,ttype=ttype, 
   2356                                         cache=False, fft_pad_factor=fft_pad_factor,
   2357                                         zero_empty_pol=True, verbose=verbose)

~/.local/lib/python3.10/site-packages/ehtim/image.py in observe_same_nonoise(self, obs, sgrscat, ttype, cache, fft_pad_factor, zero_empty_pol, verbose)
   2270         # Extract uv datasample
   2271         uv = obsh.recarr_to_ndarr(obsdata[['u', 'v']], 'f8')
-> 2272         data = simobs.sample_vis(self, uv, sgrscat=sgrscat, polrep_obs=obs.polrep,
   2273                                  ttype=ttype, cache=cache, fft_pad_factor=fft_pad_factor,
   2274                                  zero_empty_pol=zero_empty_pol, verbose=verbose)

~/.local/lib/python3.10/site-packages/ehtim/observing/obs_simulate.py in sample_vis(im_org, uv, sgrscat, polrep_obs, ttype, cache, fft_pad_factor, zero_empty_pol, verbose)
    299         uvdim = len(uv)
    300         if (im.xdim % 2 or im.ydim % 2):
--> 301             raise Exception("NFFT doesn't work with odd image dimensions!")
    302 
    303         npad = fft_pad_factor * np.max((im.xdim, im.ydim))

Exception: NFFT doesn't work with odd image dimensions!

似乎NFFT与.fits不起作用，具有奇数图像尺寸。但是，我的.fit image只有二维形状（63,63），可以通过使用Astropy来显示：

import matplotlib.pyplot as plt
from astropy.utils.data import get_pkg_data_filename
from astropy.io import fits

image_file = get_pkg_data_filename('disk_SchildWH.fits')


fits.info(image_file)

这给出了

Filename: disk_SchildWH.fits
No.    Name      Ver    Type      Cards   Dimensions   Format
  0  PRIMARY       1 PrimaryHDU       8   (63, 63)   float64   

此拟合文件的尺寸均匀，并且所有内容都可以正常工作，但是> > EHTIM无法产生图像。也许有人知道为什么会发生？在这里，我添加了代码中使用的文件（eht txt array and .fits file）：

https://drive.google.com/file/d/1bhpj52w4qhs5pv--xd7_sjrnkb_yr4or/view?usp = sharing？usp = sharing

？ /file/d/d/1fpqesn0aigila5ey1hgptdhlzhhixh2_/view？usp =共享“ rel =” nofollow noreferrer“> https://drive.google.com/file.com/file.com/file/file/d/1fpqesn0aigila5ey1 ey1hgptdhlzhlzhh2_h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2h2spr.shring us