def check_geweke(self, setting_idx, hypparam_dict):
import os
from matplotlib import pyplot as plt
plt.ioff()
fig = plt.figure()
figpath = self.geweke_figure_filepath(setting_idx)
mkdir(os.path.dirname(figpath))
nsamples, data_size, ntrials = self.geweke_nsamples, \
self.geweke_data_size, self.geweke_ntrials
d = self.distribution_class(**hypparam_dict)
sample_dim = np.atleast_1d(self.geweke_statistics(
d, d.rvs(size=10))).shape[0]
num_statistic_fails = 0
for trial in range(ntrials):
# collect forward-generated statistics
forward_statistics = np.squeeze(np.empty((nsamples, sample_dim)))
for i in range(nsamples):
d = self.distribution_class(**hypparam_dict)
data = d.rvs(size=data_size)
forward_statistics[i] = self.geweke_statistics(d, data)
# collect gibbs-generated statistics
gibbs_statistics = np.squeeze(np.empty((nsamples, sample_dim)))
d = self.distribution_class(**hypparam_dict)
data = d.rvs(size=data_size)
for i in range(nsamples):
d.resample(data, **self.geweke_resample_kwargs)
data = d.rvs(size=data_size)
gibbs_statistics[i] = self.geweke_statistics(d, data)
testing.populations_eq_quantile_plot(forward_statistics,
gibbs_statistics,
fig=fig)
try:
sl = self.geweke_numerical_slice(d, setting_idx)
testing.assert_populations_eq_moments(forward_statistics[...,
sl],
gibbs_statistics[...,
sl],
pval=self.geweke_pval)
except AssertionError:
datapath = os.path.join(
os.path.dirname(__file__), 'figures',
self.__class__.__name__,
'setting_%d_trial_%d.npz' % (setting_idx, trial))
np.savez(datapath,
fwd=forward_statistics,
gibbs=gibbs_statistics)
example_violating_means = forward_statistics.mean(
0), gibbs_statistics.mean(0)
num_statistic_fails += 1
plt.savefig(figpath)
assert num_statistic_fails <= self.geweke_num_statistic_fails_to_tolerate, \
'Geweke MAY have failed, check FIGURES in %s (e.g. %s vs %s)' \
% ((os.path.dirname(figpath),) + example_violating_means)
def test_gaussian():
prior_data = 2 * np.random.randn(5, 2) + np.array([1., 3.])
a = Gaussian().empirical_bayes(prior_data)
# data = a.rvs(10)
gibbs_statistics = []
for itr in range(20000):
a.resample()
# a.resample(data)
gibbs_statistics.append(a.mu)
gibbs_statistics = np.array(gibbs_statistics)
b = AnnealedGaussianModel().empirical_bayes(prior_data)
# b.add_data(data)
pt = ParallelTempering(b, [5.])
pt_samples = pt.run(20000, 1)
pt_statistics = np.array([m.mu for m in pt_samples])
fig = plt.figure()
testing.populations_eq_quantile_plot(gibbs_statistics,
pt_statistics,
fig=fig)
plt.savefig('gaussian_test.png')
testing.assert_populations_eq_moments(gibbs_statistics,pt_statistics), \
'Annealing MAY have failed, check FIGURES'
def test_gaussian():
prior_data = 2*np.random.randn(5,2) + np.array([1.,3.])
a = Gaussian().empirical_bayes(prior_data)
# data = a.rvs(10)
gibbs_statistics = []
for itr in range(20000):
a.resample()
# a.resample(data)
gibbs_statistics.append(a.mu)
gibbs_statistics = np.array(gibbs_statistics)
b = AnnealedGaussianModel().empirical_bayes(prior_data)
# b.add_data(data)
pt = ParallelTempering(b,[5.])
pt_samples = pt.run(20000,1)
pt_statistics = np.array([m.mu for m in pt_samples])
fig = plt.figure()
testing.populations_eq_quantile_plot(gibbs_statistics,pt_statistics,fig=fig)
plt.savefig('gaussian_test.png')
testing.assert_populations_eq_moments(gibbs_statistics,pt_statistics), \
'Annealing MAY have failed, check FIGURES'
def discrete_geweke_test(fig):
Nstates = 2
Nemissions = 2
alpha = 3.
init_state_concentration=3.
T = 10
num_iter = 10000
num_checks = 10
obs_distns = [distributions.Categorical(K=Nemissions,alpha_0=1.)
for _ in range(Nstates)]
hmm = models.HMM(
alpha=alpha,init_state_concentration=init_state_concentration,
obs_distns=obs_distns)
# generate state sequences and parameters from the prior
prior_stateseqs = []
prior_weights = []
for itr in range(num_iter):
hmm.resample_model() # sample parameters from the prior
_, stateseq = hmm.generate(T,keep=False)
prior_stateseqs.append(stateseq)
prior_weights.append(hmm.obs_distns[0].weights)
prior_stateseqs = np.array(prior_stateseqs)
prior_weights = np.array(prior_weights)
# generate state sequences and parameters using Gibbs
hmm.generate(T,keep=True)
s = hmm.states_list[0]
gibbs_stateseqs = []
gibbs_weights = []
for itr in range(num_iter):
s.data = None
hmm._generate_obs(s) # resamples data given state sequence, obs params
hmm.resample_model() # resamples everything else as usual
gibbs_stateseqs.append(s.stateseq)
gibbs_weights.append(hmm.obs_distns[0].weights)
gibbs_stateseqs = np.array(gibbs_stateseqs)
gibbs_weights = np.array(gibbs_weights)
# test that they look similar by checking probability of co-assignment
time_indices = np.arange(T)
for itr in range(num_checks):
i,j = np.random.choice(time_indices,replace=False,size=2)
prior_prob_of_coassignment = (prior_stateseqs[:,i] == prior_stateseqs[:,j]).std()
gibbs_prob_of_coassignment = (gibbs_stateseqs[:,i] == gibbs_stateseqs[:,j]).std()
assert np.isclose(
prior_prob_of_coassignment,gibbs_prob_of_coassignment,
rtol=0.025,atol=0.025,
)
# test that they look similar by checking parameters
testing.populations_eq_quantile_plot(prior_weights,gibbs_weights,fig=fig)
figpath = os.path.join(figure_dir_path,'discrete_geweke_test_weights.pdf')
plt.savefig(figpath)
def check_geweke(self, setting_idx, hypparam_dict):
import os
from matplotlib import pyplot as plt
plt.ioff()
fig = plt.figure()
figpath = self.geweke_figure_filepath(setting_idx)
mkdir(os.path.dirname(figpath))
nsamples, data_size, ntrials = self.geweke_nsamples, self.geweke_data_size, self.geweke_ntrials
d = self.distribution_class(**hypparam_dict)
sample_dim = np.atleast_1d(self.geweke_statistics(d, d.rvs(size=10))).shape[0]
num_statistic_fails = 0
for trial in xrange(ntrials):
# collect forward-generated statistics
forward_statistics = np.squeeze(np.empty((nsamples, sample_dim)))
for i in xrange(nsamples):
d = self.distribution_class(**hypparam_dict)
data = d.rvs(size=data_size)
forward_statistics[i] = self.geweke_statistics(d, data)
# collect gibbs-generated statistics
gibbs_statistics = np.squeeze(np.empty((nsamples, sample_dim)))
d = self.distribution_class(**hypparam_dict)
data = d.rvs(size=data_size)
for i in xrange(nsamples):
d.resample(data, **self.geweke_resample_kwargs)
data = d.rvs(size=data_size)
gibbs_statistics[i] = self.geweke_statistics(d, data)
testing.populations_eq_quantile_plot(forward_statistics, gibbs_statistics, fig=fig)
try:
sl = self.geweke_numerical_slice(d, setting_idx)
testing.assert_populations_eq_moments(
forward_statistics[..., sl], gibbs_statistics[..., sl], pval=self.geweke_pval
)
except AssertionError:
datapath = os.path.join(
os.path.dirname(__file__),
"figures",
self.__class__.__name__,
"setting_%d_trial_%d.npz" % (setting_idx, trial),
)
np.savez(datapath, fwd=forward_statistics, gibbs=gibbs_statistics)
example_violating_means = forward_statistics.mean(0), gibbs_statistics.mean(0)
num_statistic_fails += 1
plt.savefig(figpath)
assert num_statistic_fails <= self.geweke_num_statistic_fails_to_tolerate, (
"Geweke MAY have failed, check FIGURES in %s (e.g. %s vs %s)"
% ((os.path.dirname(figpath),) + example_violating_means)
)
def discrete_geweke_test(fig):
Nstates = 2
Nemissions = 2
alpha = 3.
init_state_concentration = 3.
T = 10
num_iter = 10000
num_checks = 10
obs_distns = [
distributions.Categorical(K=Nemissions, alpha_0=1.)
for _ in range(Nstates)
]
hmm = models.HMM(alpha=alpha,
init_state_concentration=init_state_concentration,
obs_distns=obs_distns)
# generate state sequences and parameters from the prior
prior_stateseqs = []
prior_weights = []
for itr in range(num_iter):
hmm.resample_model() # sample parameters from the prior
_, stateseq = hmm.generate(T, keep=False)
prior_stateseqs.append(stateseq)
prior_weights.append(hmm.obs_distns[0].weights)
prior_stateseqs = np.array(prior_stateseqs)
prior_weights = np.array(prior_weights)
# generate state sequences and parameters using Gibbs
hmm.generate(T, keep=True)
s = hmm.states_list[0]
gibbs_stateseqs = []
gibbs_weights = []
for itr in range(num_iter):
s.data = None
hmm._generate_obs(s) # resamples data given state sequence, obs params
hmm.resample_model() # resamples everything else as usual
gibbs_stateseqs.append(s.stateseq)
gibbs_weights.append(hmm.obs_distns[0].weights)
gibbs_stateseqs = np.array(gibbs_stateseqs)
gibbs_weights = np.array(gibbs_weights)
# test that they look similar by checking probability of co-assignment
time_indices = np.arange(T)
for itr in range(num_checks):
i, j = np.random.choice(time_indices, replace=False, size=2)
prior_prob_of_coassignment = (
prior_stateseqs[:, i] == prior_stateseqs[:, j]).std()
gibbs_prob_of_coassignment = (
gibbs_stateseqs[:, i] == gibbs_stateseqs[:, j]).std()
assert np.isclose(
prior_prob_of_coassignment,
gibbs_prob_of_coassignment,
rtol=0.025,
atol=0.025,
)
# test that they look similar by checking parameters
testing.populations_eq_quantile_plot(prior_weights, gibbs_weights, fig=fig)
figpath = os.path.join(figure_dir_path, 'discrete_geweke_test_weights.pdf')
plt.savefig(figpath)