def pymask_grid(input_data,
ngrid=40,
pa_prior=[0, 360],
sep_prior=[0, 100],
cr_prior=[1, 150],
err_scale=1.,
extra_error_cp=0.,
ncore=1,
verbose=False):
cpo = pymask.cpo(input_data)
like_grid = pymask.coarse_grid(cpo,
nsep=ngrid,
nth=ngrid,
ncon=ngrid,
thmin=pa_prior[0],
thmax=pa_prior[1],
smin=sep_prior[0],
smax=sep_prior[1],
cmin=cr_prior[0],
cmax=cr_prior[1],
threads=ncore,
err_scale=err_scale,
extra_error=extra_error_cp,
verbose=verbose)
return like_grid
def pymask_cr_limit(input_data,
nsim=100,
err_scale=1,
extra_error_cp=0,
ncore=1,
cmax=500,
nsep=60,
ncrat=60,
nth=30,
smin=20,
smax=250,
cmin=1.0001):
cpo = pymask.cpo(input_data)
lims_data = pymask.detec_limits(cpo,
threads=ncore,
nsim=nsim,
nsep=nsep,
ncon=ncrat,
nth=nth,
smax=smax,
cmax=cmax,
cmin=cmin,
smin=smin,
err_scale=err_scale,
extra_error=extra_error_cp)
limits = lims_data['limits']
seps = lims_data['seps']
crats = lims_data['cons']
crat_limits = 0 * seps
# Loop through seps and find the highest contrast ratio that would be detectable
for sep_ix in range(len(seps)):
would_detec = limits[:, sep_ix] >= 0.9973
if np.sum(would_detec) >= 1:
threesig_lim = np.max(crats[would_detec])
crat_limits[sep_ix] = threesig_lim # fivesig_lim
else:
crat_limits[sep_ix] = 1.
con_limits = 2.5 * np.log10(crat_limits)
plt.figure()
plt.plot(seps, con_limits)
plt.xlabel('Separation [mas]')
plt.ylabel('$\Delta \mathrm{Mag}_{3\sigma}$')
plt.title("PYMASK: flux ratio for 3$\sigma$ detection")
plt.ylim(plt.ylim()[1], plt.ylim()[0]) # -- rreverse plot
plt.tight_layout()
res = {'r': seps, 'cr_limit': con_limits, 'lims_data': lims_data}
return res
def pymask_mcmc(input_data,
initial_guess,
niters=1000,
pa_prior=[0, 360],
sep_prior=None,
cr_prior=None,
err_scale=1,
extra_error_cp=0,
ncore=1,
burn_in=500,
walkers=100,
display=True,
verbose=True):
cpo = pymask.cpo(input_data)
hammer_data = pymask.hammer(cpo,
ivar=initial_guess,
niters=niters,
model='constant',
nwalcps=walkers,
sep_prior=sep_prior,
pa_prior=pa_prior,
crat_prior=cr_prior,
err_scale=err_scale,
extra_error=extra_error_cp,
plot=display,
burn_in=burn_in,
threads=ncore)
res_corner = hammer_data[1]
chain = hammer_data[0]['chain']
dm = 2.5 * np.log10(res_corner['cr'])
dmm = 2.5 * np.log10(res_corner['cr'] - res_corner['delcrm'])
dmp = 2.5 * np.log10(res_corner['cr'] + res_corner['delcrp'])
e_dmm = abs(dm - dmm)
e_dmp = abs(dm - dmp)
if verbose:
print('MCMC estimation')
print('---------------')
print(
'Separation = %2.1f +%2.1f/-%2.1f mas' %
(res_corner['sep'], res_corner['delsepp'], res_corner['delsepm']))
print('PA = %2.1f +%2.1f/-%2.1f deg' %
(res_corner['pa'], res_corner['delpap'], res_corner['delpam']))
print('Contrast Ratio = %2.1f +%2.1f/-%2.1f' %
(res_corner['cr'], res_corner['delcrp'], res_corner['delcrm']))
print('dm = %2.2f +%2.2f/-%2.2f mag' % (dm, e_dmp, e_dmm))
chain_sep = chain[:, :, 0].T
chain_th = chain[:, :, 1].T
chain_cr = chain[:, :, 2].T
if display:
sep = res_corner['sep']
pa = res_corner['pa']
cr = res_corner['cr']
plt.figure(figsize=(5, 7))
plt.subplot(3, 1, 1)
plt.plot(chain_sep, color='grey', alpha=.5)
plt.plot(len(chain_sep), sep, marker='*', color='#0085ca', zorder=1e3)
plt.ylabel('Separation [mas]')
plt.subplot(3, 1, 2)
plt.plot(chain_th, color='grey', alpha=.5)
plt.plot(len(chain_sep), pa, marker='*', color='#0085ca', zorder=1e3)
plt.ylabel('PA [deg]')
plt.subplot(3, 1, 3)
plt.plot(chain_cr, color='grey', alpha=.2)
plt.plot(len(chain_sep), cr, marker='*', color='#0085ca', zorder=1e3)
plt.xlabel('Step')
plt.ylabel('CR')
plt.tight_layout()
plt.show(block=False)
res = {
'best': {
'model': 'binary',
'dm': dm,
'theta': res_corner['pa'],
'sep': res_corner['sep'],
'x0': 0,
'y0': 0
},
'uncer': {
'dm_p': e_dmp,
'dm_m': e_dmm,
'theta_p': res_corner['delpap'],
'theta_m': res_corner['delpam'],
'sep_p': res_corner['delsepp'],
'sep_m': res_corner['delsepm']
},
}
return res