c0 = np.random.normal(true_mu_c0, true_sigma_c0)
c1 = np.random.uniform(-1, 1)
injection_parameters = dict(c0=c0, c1=c1)
data = model(x, **injection_parameters) + np.random.normal(0, sigma, N)
line = ax1.plot(x, data, "-x", label=labels[i])
likelihood = GaussianLikelihood(x, data, model, sigma)
priors = dict(c0=Uniform(-10, 10, "c0"), c1=Uniform(-10, 10, "c1"))
result = run_sampler(
likelihood=likelihood,
priors=priors,
sampler="nestle",
nlive=1000,
outdir=outdir,
verbose=False,
label="individual_{}".format(i),
save=False,
injection_parameters=injection_parameters,
)
ax2.hist(
result.posterior.c0,
color=line[0].get_color(),
density=True,
alpha=0.5,
label=labels[i],
)
results.append(result)
ax1.set_xlabel("x")
n_dim = 3
label = "dirichlet_"
priors = DirichletPriorDict(n_dim=n_dim, label=label)
injection_parameters = dict(
dirichlet_0=1 / 3,
dirichlet_1=1 / 3,
dirichlet_2=1 / 3,
)
data = [injection_parameters[label + str(ii)] * 1000 for ii in range(n_dim)]
likelihood = Multinomial(data=data, n_dimensions=n_dim, label=label)
result = run_sampler(likelihood=likelihood,
priors=priors,
nlive=100,
walks=10,
label="multinomial",
injection_parameters=injection_parameters)
result.posterior[label + str(n_dim - 1)] = 1 - np.sum(
[result.posterior[key] for key in priors], axis=0)
result.plot_corner(parameters=injection_parameters)
samples = priors.sample(10000)
samples[label +
str(n_dim - 1)] = 1 - np.sum([samples[key] for key in samples], axis=0)
result.posterior = pd.DataFrame(samples)
result.plot_corner(parameters=[key for key in samples],
filename="outdir/dirichlet_prior_corner.png")
for i in range(Nevents):
c0 = np.random.normal(true_mu_c0, true_sigma_c0)
c1 = np.random.uniform(-1, 1)
injection_parameters = dict(c0=c0, c1=c1)
data = model(x, **injection_parameters) + np.random.normal(0, sigma, N)
line = ax1.plot(x, data, '-x', label=labels[i])
likelihood = GaussianLikelihood(x, data, model, sigma)
priors = dict(c0=Uniform(-10, 10, 'c0'), c1=Uniform(-10, 10, 'c1'))
result = run_sampler(likelihood=likelihood,
priors=priors,
sampler='nestle',
nlive=1000,
outdir=outdir,
verbose=False,
label='individual_{}'.format(i),
save=False,
injection_parameters=injection_parameters)
ax2.hist(result.posterior.c0,
color=line[0].get_color(),
normed=True,
alpha=0.5,
label=labels[i])
results.append(result)
ax1.set_xlabel('x')
ax1.set_ylabel('y(x)')
ax1.legend()
fig1.savefig('outdir/hyper_parameter_data.png')