def cornerpoints(self, **kwargs):
try:
dyplot.cornerpoints(results=self.samples.results,
labels=self.model.parameter_labels_latex,
**kwargs)
self.output.to_figure(structure=None, auto_filename="cornerpoints")
except ValueError:
pass
def test_gaussian():
logz_tol = 1
sampler = dynesty.NestedSampler(loglikelihood_gau,
prior_transform_gau,
ndim_gau,
nlive=nlive)
sampler.run_nested(print_progress=printing)
# add samples
# check continuation behavior
sampler.run_nested(dlogz=0.1, print_progress=printing)
# get errors
nerr = 2
for i in range(nerr):
sampler.reset()
sampler.run_nested(print_progress=False)
results = sampler.results
pos = results.samples
wts = np.exp(results.logwt - results.logz[-1])
mean, cov = dyfunc.mean_and_cov(pos, wts)
logz = results.logz[-1]
assert (np.abs(logz - logz_truth_gau) < logz_tol)
# check summary
res = sampler.results
res.summary()
# check plots
dyplot.runplot(sampler.results)
plt.close()
dyplot.traceplot(sampler.results)
plt.close()
dyplot.cornerpoints(sampler.results)
plt.close()
dyplot.cornerplot(sampler.results)
plt.close()
dyplot.boundplot(sampler.results,
dims=(0, 1),
it=3000,
prior_transform=prior_transform_gau,
show_live=True,
span=[(-10, 10), (-10, 10)])
plt.close()
dyplot.cornerbound(sampler.results,
it=3500,
prior_transform=prior_transform_gau,
show_live=True,
span=[(-10, 10), (-10, 10)])
plt.close()
def cornerpoints(self, **kwargs):
if conf.instance["general"]["test"]["test_mode"]:
return None
try:
dyplot.cornerpoints(results=self.samples.results,
labels=self.model.parameter_labels_latex,
**kwargs)
self.output.to_figure(structure=None, auto_filename="cornerpoints")
except ValueError:
pass
self.mat_plot_1d.figure.close()
def main(args=None):
if args.results == "none":
ndim = len(JOINT_PRIOR)
date = datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
image_positions = pd.read_csv(args.image_positions)
pprint(image_positions)
# A=0, B=1, C=2, D=3
x_image = image_positions["theta_x"].to_numpy()
y_image = image_positions["theta_y"].to_numpy()
quad_model = QuadPseudoNIELensModel(x_image, y_image, args.plate_scale)
time_delays = pd.read_csv(args.time_delays)
pprint(time_delays)
# Expected: DA=0, DB=1, DC=2 (index)
sigma_t = time_delays["sigma"].to_numpy() # units of days
delta_t = time_delays["delta_t"].to_numpy()
joint_loglikelihood = joint_loglikelihood_func(quad_model, delta_t,
sigma_t)
with Pool(args.nworkers) as pool:
sampler = DynamicNestedSampler(joint_loglikelihood,
joint_prior_transform,
ndim,
pool=pool,
nlive=args.nlive,
queue_size=args.nworkers)
sampler.run_nested()
res = sampler.results
# save results
with open(f"../results/joint_result_{date}.p", "wb") as f:
pickle.dump(res, f)
else: # we just want to plot the result from an older run
with open(args.results, "rb") as f:
res = pickle.load(f)
ndim = res.samples.shape[1]
if args.plot_results:
# trace plot
fig, axs = dyplot.traceplot(
res,
show_titles=True,
trace_cmap='plasma',
connect=True,
connect_highlight=range(5),
labels=LABELS,
)
fig.tight_layout(pad=2.0)
fig.savefig("../figures/joint_inference_trace_plot.png",
bbox_inches="tight")
# corner points plot
fig, axes = plt.subplots(ndim - 1, ndim - 1, figsize=(15, 15))
axes.reshape([ndim - 1, ndim - 1])
fg, ax = dyplot.cornerpoints(res,
cmap='plasma',
kde=False,
fig=(fig, axes),
labels=LABELS)
fg.savefig("../figures/joint_inference_cornerpoints.png",
bbox_inches="tight")
# corner plot
fig, axes = plt.subplots(ndim, ndim, figsize=(15, 15))
axes.reshape([ndim, ndim])
fg, ax = dyplot.cornerplot(res,
fig=(fig, axes),
color="b",
labels=LABELS,
show_titles=True)
fg.savefig("../figures/joint_inference_corner_plot.png",
bbox_inches="tight")
#### marginalized posterior #####
Ddt = res.samples[:, 0] # histogram of Ddt
weights = np.exp(
res['logwt'] -
res['logz'][-1]) # posterior probability (Bayes theorem)
# eliminate outliers with low and high + estimate confidance interval
low, fifth, median, ninety_fifth, high = weighted_quantile(
Ddt, [0.0001, 0.05, 0.5, 0.95, 0.9999], weights)
error_plus = ninety_fifth - median # error estimate with 5th percentile
error_minus = median - fifth # error estimate with 95th percentile
good = (Ddt > low) & (Ddt < high) # remove outliers
Ddt = Ddt[good]
weights = weights[good]
plt.figure(figsize=(8, 8))
plt.hist(Ddt, bins=100, weights=weights)
plt.title(r"$D_{\Delta t}$=%.2f$^{+%.2f}_{-%.2f}$" %
(median, error_plus, error_minus))
plt.xlabel(r"$D_{\Delta t}$")
plt.savefig("../figures/marginalized_posterior_Ddt.png")
# assume a flat LambdCDM model (with negligible radiation)
# We need to model kappa_ext for this step
def integrand(z):
return 1 / np.sqrt(args.omega_m * (1 + z)**3 + args.omega_l)
Dd = quad(integrand, 0, args.z_lens)[0] / (1 + args.z_lens)
Ds = quad(integrand, 0, args.z_source)[0] / (1 + args.z_source)
Dds = quad(integrand, args.z_lens,
args.z_source)[0] / (1 + args.z_source)
factor = (1 + args.z_lens) * Ds * Dd / Dds
H0 = (c * factor / Ddt / u.Mpc).to(u.km / u.s / u.Mpc).value
plt.figure(figsize=(8, 8))
fifth, median, ninety_fifth = weighted_quantile(
H0, [0.05, 0.5, 0.95], weights)
error_plus = ninety_fifth - median
error_minus = median - fifth
plt.hist(H0, bins=100, weights=weights)
plt.title(r"$H_0$=%.2f$^{+%.2f}_{-%.2f}$" %
(median, error_plus, error_minus))
plt.xlabel(r"$H_0$ [km s$^{-1}$ Mpc$^{-1}$]")
plt.savefig("../figures/marginalized_posterior_H0.png")
def test_gaussian(dynamic, periodic, ndim, bound):
rstate = get_rstate()
g = Gaussian(ndim=ndim)
if periodic:
periodic = [0]
else:
periodic = None
if dynamic:
sampler = dynesty.DynamicNestedSampler(g.loglikelihood,
g.prior_transform,
g.ndim,
nlive=nlive,
rstate=rstate,
periodic=periodic,
bound=bound)
else:
sampler = dynesty.NestedSampler(g.loglikelihood,
g.prior_transform,
g.ndim,
nlive=nlive,
rstate=rstate,
periodic=periodic,
bound=bound)
sampler.run_nested(print_progress=printing)
results = sampler.results
# check plots
dyplot.runplot(results)
dyplot.runplot(results, span=[(0., 10.), 0.001, 0.2, (5., 6.)])
dyplot.runplot(results, logplot=True)
dyplot.runplot(results,
fig=(plt.gcf(), plt.gcf().axes),
max_x_ticks=0,
max_y_ticks=0)
plt.close()
dyplot.traceplot(results)
dyplot.traceplot(results, smooth=[10] * ndim)
dyplot.traceplot(results, connect=True)
dyplot.traceplot(results,
fig=(plt.gcf(), plt.gcf().axes),
show_titles=True,
truths=np.zeros(ndim),
verbose=True,
max_n_ticks=0)
plt.close()
truths = np.zeros(ndim)
truths[0] = -.1
span = [[-10, 10]] * ndim
if ndim > 1:
truths[1] = .1
span[1] = .9
dyplot.cornerplot(results, show_titles=True, truths=truths)
dyplot.cornerplot(results,
smooth=10,
verbose=True,
hist2d_kwargs=dict(plot_datapoints=True),
max_n_ticks=0)
plt.close()
if ndim != 1:
# cornerbound
dyplot.cornerbound(results,
it=500,
prior_transform=g.prior_transform,
show_live=True,
span=span)
dyplot.cornerbound(results,
it=500,
show_live=True,
span=span,
fig=(plt.gcf(), plt.gcf().axes))
plt.close()
# boundplot
dyplot.boundplot(results,
dims=(0, 1)[:min(ndim, 2)],
it=1000,
prior_transform=g.prior_transform,
show_live=True,
span=span)
dyplot.boundplot(results, dims=(0, 1)[:min(ndim, 2)], it=1000)
plt.close()
# cornerpoints
dyplot.cornerpoints(results)
plt.close()
dyplot.cornerpoints(results, span=span, truths=truths, max_n_ticks=0)
plt.close()
def plot(n, alpha, beta, dlogz_val=.1, interloper=False, d=1):
x_k = np.round(generator(n, alpha, beta, d), 2) #round up
#append values if interloper is true
if (interloper == True):
interloper_x_k = np.round(generator(n, alpha + 3, beta, d), 2)
x_k = np.append(x_k, interloper_x_k)
#feed parameters for alpha and beta
def lighthouse_logl(params):
return log_likelihood(x_k, params[0], params[1])
#prior transform, using documentation for range -1000, 1000
def prior_transform(u):
return [2000 * u[0] - 1000, 1000 * u[1]]
ndim = 2
sampler = dynesty.NestedSampler(lighthouse_logl,
prior_transform,
ndim,
bound='single',
nlive=200)
sampler.run_nested(dlogz=dlogz_val, print_progress=False)
results = sampler.results
#runplot
dyplot.runplot(results)
#plt.savefig('int_run_plot' + str(dlogz_val)+".png", dpi =300)
plt.show()
#cornerpoints plot
fig = plt.subplots(1, 1, figsize=(10, 10))
dyplot.cornerpoints(results,
fig=fig,
cmap='plasma',
truths=np.zeros(ndim),
kde=False)
fig[1].set_ylabel('$\\beta$')
fig[1].set_xlabel('$\\alpha$')
plt.tight_layout()
plt.xlim(-10, 10)
plt.ylim(0, 10)
plt.savefig('int_corner1' + str(dlogz_val) + ".png", dpi=300)
plt.show()
#traceplot
fig = plt.subplots(2, 2, figsize=(15, 10))
dyplot.traceplot(results,
fig=fig,
truth_color='black',
trace_cmap='viridis',
connect=True,
connect_highlight=range(5),
show_titles=True)
fig[1][1, 1].set_xlabel('$\\beta$')
fig[1][0, 1].set_xlabel('$\\alpha$')
fig[1][1, 0].set_ylabel('$\\beta$')
fig[1][0, 0].set_ylabel('$\\alpha$')
plt.tight_layout()
#plt.savefig('int_trace' + str(dlogz_val)+".png", dpi =300)
plt.show()
print(results.samples[-1])
# check summary
sys.stderr.write('\nResults\n')
res = sampler.results
res.summary()
# check plots
sys.stderr.write('\nPlotting\n')
sys.stderr.write('Summary/Run Plot\n')
dyplot.runplot(sampler.results)
plt.close()
sys.stderr.write('Trace Plot\n')
dyplot.traceplot(sampler.results)
plt.close()
sys.stderr.write('Sub-Corner Plot (Points)\n')
dyplot.cornerpoints(sampler.results)
plt.close()
sys.stderr.write('Corner Plot (Contours)\n')
dyplot.cornerplot(sampler.results)
plt.close()
sys.stderr.write('2-D Bound Plot\n')
dyplot.boundplot(sampler.results,
dims=(0, 1),
it=3000,
prior_transform=prior_transform,
show_live=True,
span=[(-10, 10), (-10, 10)])
plt.close()
sys.stderr.write('Sub-Corner Plot (Bounds)\n')
dyplot.cornerbound(sampler.results,
it=3500,
res = sampler.sampler.results
else:
res = sampler.results
# Make the traceplots and the bound plots.
plot_status(res, labels=labels, periodic=periodic)
# Generate a plot of the weighted samples.
fig, ax = plt.subplots(ndim - 1, ndim - 1, figsize=(10, 10))
dyplot.cornerpoints(res,
cmap="plasma",
kde=False,
fig=(fig, ax),
labels=labels)
fig.savefig("cornerpoints.png")
# Generate a corner plot from Dynesty.
fig, ax = plt.subplots(ndim, ndim, figsize=(15, 15))
dyplot.cornerplot(res, color="blue", show_titles=True, max_n_ticks=3, \
quantiles=None, fig=(fig, ax), labels=labels)
fig.savefig("cornerplot.png")
# Convert the results to a more traditional set of samples that you would