def test__unconverged_sample_size__uses_value_unless_fewer_samples(self):
model = af.ModelMapper(mock_class_1=MockClassx4)
log_likelihood_list = 4 * [0.0] + [1.0]
weight_list = 4 * [0.0] + [1.0]
samples = MockSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=5 * [[]],
log_likelihood_list=log_likelihood_list,
log_prior_list=[1.0, 1.0, 1.0, 1.0, 1.0],
weight_list=weight_list,
),
unconverged_sample_size=2,
)
assert samples.pdf_converged is False
assert samples.unconverged_sample_size == 2
samples = MockSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=5 * [[]],
log_likelihood_list=log_likelihood_list,
log_prior_list=[1.0, 1.0, 1.0, 1.0, 1.0],
weight_list=weight_list,
),
unconverged_sample_size=6,
)
assert samples.pdf_converged is False
assert samples.unconverged_sample_size == 5
def test__search_summary_to_file(model):
file_search_summary = path.join(text_path, "search.summary")
parameters = [[1.0, 2.0], [1.2, 2.2]]
log_likelihood_list = [1.0, 0.0]
samples = MockSamples(
model=model,
samples=Sample.from_lists(
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=[0.0, 0.0],
weight_list=log_likelihood_list,
model=model
),
time=None,
)
text_util.search_summary_to_file(samples=samples, log_likelihood_function_time=1.0, filename=file_search_summary)
results = open(file_search_summary)
lines = results.readlines()
assert lines[0] == "Total Samples = 2\n"
results.close()
samples = MockNestSamples(
model=model,
samples=Sample.from_lists(
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list + [2.0],
log_prior_list=[1.0, 1.0],
weight_list=log_likelihood_list,
model=model
),
total_samples=10,
time=2,
number_live_points=1,
log_evidence=1.0,
)
text_util.search_summary_to_file(samples=samples, log_likelihood_function_time=1.0, filename=file_search_summary)
results = open(file_search_summary)
lines = results.readlines()
assert lines[0] == "Total Samples = 10\n"
assert lines[1] == "Total Accepted Samples = 2\n"
assert lines[2] == "Acceptance Ratio = 0.2\n"
assert lines[3] == "Time To Run = 2\n"
assert lines[4] == "Log Likelihood Function Evaluation Time (seconds) = 1.0"
results.close()
def test__offset_vector_from_input_vector(self):
model = af.ModelMapper(mock_class_1=MockClassx4)
parameters = [
[1.1, 2.1, 3.1, 4.1],
[1.0, 2.0, 3.0, 4.0],
[1.0, 2.0, 3.0, 4.0],
[1.0, 2.0, 3.0, 4.0],
[1.0, 2.0, 3.0, 4.1],
]
weight_list = [0.3, 0.2, 0.2, 0.2, 0.1]
log_likelihood_list = list(
map(lambda weight: 10.0 * weight, weight_list))
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
offset_values = samples.offset_vector_from_input_vector(
input_vector=[1.0, 1.0, 2.0, 3.0])
assert offset_values == pytest.approx([0.0, 1.0, 1.0, 1.025], 1.0e-4)
def test__unconverged__median_pdf_vector(self):
parameters = [
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.1, 2.1],
[0.9, 1.9],
]
log_likelihood_list = 9 * [0.0] + [1.0]
weight_list = 9 * [0.0] + [1.0]
model = af.ModelMapper(mock_class=MockClassx2)
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
assert samples.pdf_converged is False
median_pdf_vector = samples.median_pdf_vector
assert median_pdf_vector[0] == pytest.approx(0.9, 1.0e-4)
assert median_pdf_vector[1] == pytest.approx(1.9, 1.0e-4)
def samples(self):
"""
Create a `Samples` object from this non-linear search's output files on the hard-disk and model.
Parameters
----------
model
The model which generates instances for different points in parameter space. This maps the points from unit
cube values to physical values via the priors.
paths : af.Paths
Manages all paths, e.g. where the search outputs are stored, the `NonLinearSearch` chains,
etc.
"""
if self._samples is not None:
return self._samples
parameter_lists = self.zeus_sampler.get_chain(flat=True).tolist()
log_prior_list = [
sum(self.model.log_prior_list_from_vector(vector=vector)) for vector in parameter_lists
]
log_posterior_list = self.zeus_sampler.get_log_prob(flat=True).tolist()
log_likelihood_list = [log_posterior - log_prior for log_posterior, log_prior in zip(log_posterior_list, log_prior_list)]
weight_list = len(log_likelihood_list) * [1.0]
self._samples = Sample.from_lists(
model=self.model,
parameter_lists=parameter_lists,
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list
)
return self._samples
def test__instance_from_sample_index(self):
model = af.ModelMapper(mock_class=MockClassx4)
parameters = [
[1.0, 2.0, 3.0, 4.0],
[5.0, 6.0, 7.0, 8.0],
[1.0, 2.0, 3.0, 4.0],
[1.0, 2.0, 3.0, 4.0],
[1.1, 2.1, 3.1, 4.1],
]
samples = MockSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[0.0, 0.0, 0.0, 0.0, 0.0],
log_prior_list=[0.0, 0.0, 0.0, 0.0, 0.0],
weight_list=[1.0, 1.0, 1.0, 1.0, 1.0],
))
instance = samples.instance_from_sample_index(sample_index=0)
assert instance.mock_class.one == 1.0
assert instance.mock_class.two == 2.0
assert instance.mock_class.three == 3.0
assert instance.mock_class.four == 4.0
instance = samples.instance_from_sample_index(sample_index=1)
assert instance.mock_class.one == 5.0
assert instance.mock_class.two == 6.0
assert instance.mock_class.three == 7.0
assert instance.mock_class.four == 8.0
def test__gaussian_priors(self):
parameters = [
[1.0, 2.0, 3.0, 4.0],
[1.0, 2.0, 3.0, 4.1],
[1.0, 2.0, 3.0, 4.1],
[0.88, 1.88, 2.88, 3.88],
[1.12, 2.12, 3.12, 4.32],
]
model = af.ModelMapper(mock_class=MockClassx4)
samples = MockSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[10.0, 0.0, 0.0, 0.0, 0.0],
log_prior_list=[0.0, 0.0, 0.0, 0.0, 0.0],
weight_list=[1.0, 1.0, 1.0, 1.0, 1.0],
))
gaussian_priors = samples.gaussian_priors_at_sigma(sigma=1.0)
assert gaussian_priors[0][0] == 1.0
assert gaussian_priors[1][0] == 2.0
assert gaussian_priors[2][0] == 3.0
assert gaussian_priors[3][0] == 4.0
assert gaussian_priors[0][1] == pytest.approx(0.12, 1.0e-4)
assert gaussian_priors[1][1] == pytest.approx(0.12, 1.0e-4)
assert gaussian_priors[2][1] == pytest.approx(0.12, 1.0e-4)
assert gaussian_priors[3][1] == pytest.approx(0.32, 1.0e-4)
def samples(self):
"""
Create a `Samples` object from this non-linear search's output files on the hard-disk and model.
For Emcee, all quantities are extracted via the hdf5 backend of results.
Parameters
----------
model
The model which generates instances for different points in parameter space. This maps the points from unit
cube values to physical values via the priors.
paths : af.Paths
Manages all paths, e.g. where the search outputs are stored, the `NonLinearSearch` chains,
etc.
"""
if self._samples is not None:
return self._samples
parameters = self.results["weighted_samples"]["points"]
log_likelihood_list = self.results["weighted_samples"]["logl"]
log_prior_list = [
sum(self.model.log_prior_list_from_vector(vector=vector))
for vector in parameters
]
weight_list = self.results["weighted_samples"]["weights"]
self._samples = Sample.from_lists(
model=self.model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list)
return self._samples
def test__log_prior_list_and_max_log_posterior_vector_and_instance(self):
model = af.ModelMapper(mock_class_1=MockClassx4)
parameters = [
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[21.0, 22.0, 23.0, 24.0],
]
samples = MockSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[1.0, 2.0, 3.0, 0.0, 5.0],
log_prior_list=[1.0, 2.0, 3.0, 10.0, 6.0],
weight_list=[1.0, 1.0, 1.0, 1.0, 1.0],
))
assert samples.log_posterior_list == [2.0, 4.0, 6.0, 10.0, 11.0]
assert samples.max_log_posterior_vector == [21.0, 22.0, 23.0, 24.0]
instance = samples.max_log_posterior_instance
assert instance.mock_class_1.one == 21.0
assert instance.mock_class_1.two == 22.0
assert instance.mock_class_1.three == 23.0
assert instance.mock_class_1.four == 24.0
def from_results_internal(
cls,
results_internal: np.ndarray,
log_posterior_list: np.ndarray,
model: AbstractPriorModel,
total_iterations: int,
time: Optional[float] = None,
):
"""
The `Samples` classes in **PyAutoFit** provide an interface between the results of a `NonLinearSearch` (e.g.
as files on your hard-disk) and Python.
To create a `Samples` object after an `pyswarms` model-fit the results must be converted from the
native format used by `pyswarms` (which are numpy ndarrays) to lists of values, the format used by
the **PyAutoFit** `Samples` objects.
This classmethod performs this conversion before creating a `PySwarmsSamples` object.
Parameters
----------
results_internal
The Pyswarms results in their native internal format from which the samples are computed.
log_posterior_list
The log posterior of the PySwarms accepted samples.
model
Maps input vectors of unit parameter values to physical values and model instances via priors.
total_iterations
The total number of PySwarms iterations, which cannot be estimated from the sample list (which contains
only accepted samples).
time
The time taken to perform the model-fit, which is passed around `Samples` objects for outputting
information on the overall fit.
"""
parameter_lists = [
param.tolist() for parameters in results_internal
for param in parameters
]
log_prior_list = model.log_prior_list_from(
parameter_lists=parameter_lists)
log_likelihood_list = [
lp - prior for lp, prior in zip(log_posterior_list, log_prior_list)
]
weight_list = len(log_likelihood_list) * [1.0]
sample_list = Sample.from_lists(
model=model,
parameter_lists=[
parameters.tolist()[0] for parameters in results_internal
],
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list)
return PySwarmsSamples(
model=model,
sample_list=sample_list,
total_iterations=total_iterations,
time=time,
results_internal=results_internal,
)
def samples_from(self, model):
parameter_lists = self.paths.load_object("parameter_lists")
log_prior_list = [
sum(model.log_prior_list_from_vector(vector=vector))
for vector in parameter_lists
]
log_likelihood_list = [
lp - prior for lp, prior in zip(
self.paths.load_object("log_posterior_list"), log_prior_list)
]
weight_list = len(log_likelihood_list) * [1.0]
sample_list = Sample.from_lists(
model=model,
parameter_lists=parameter_lists,
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list)
return Samples(model=model,
sample_list=sample_list,
time=self.timer.time)
def from_results_internal(
cls,
results_internal: Results,
model: AbstractPriorModel,
number_live_points: int,
unconverged_sample_size: int = 100,
time: Optional[float] = None,
):
"""
The `Samples` classes in **PyAutoFit** provide an interface between the results of a `NonLinearSearch` (e.g.
as files on your hard-disk) and Python.
To create a `Samples` object after a `dynesty` model-fit the results must be converted from the
native format used by `dynesty` to lists of values, the format used by the **PyAutoFit** `Samples` objects.
This classmethod performs this conversion before creating a `DynestySamples` object.
Parameters
----------
results_internal
The `dynesty` results in their native internal format from which the samples are computed.
model
Maps input vectors of unit parameter values to physical values and model instances via priors.
number_live_points
The number of live points used by the `dynesty` search.
unconverged_sample_size
If the samples are for a search that is yet to convergence, a reduced set of samples are used to provide
a rough estimate of the parameters. The number of samples is set by this parameter.
time
The time taken to perform the model-fit, which is passed around `Samples` objects for outputting
information on the overall fit.
"""
parameter_lists = results_internal.samples.tolist()
log_prior_list = model.log_prior_list_from(parameter_lists=parameter_lists)
log_likelihood_list = list(results_internal.logl)
try:
weight_list = list(
np.exp(np.asarray(results_internal.logwt) - results_internal.logz[-1])
)
except:
weight_list = results_internal["weights"]
sample_list = Sample.from_lists(
model=model,
parameter_lists=parameter_lists,
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list,
)
return DynestySamples(
model=model,
sample_list=sample_list,
number_live_points=number_live_points,
unconverged_sample_size=unconverged_sample_size,
time=time,
results_internal=results_internal,
)
def make_samples(model):
parameters = [[1.0, 2.0], [1.2, 2.2]]
log_likelihood_list = [1.0, 0.0]
return StoredSamples(model=model,
samples=Sample.from_lists(
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=[0.0, 0.0],
weight_list=log_likelihood_list,
model=model))
def test__samples_within_parameter_range(self, samples):
model = af.ModelMapper(mock_class_1=MockClassx4)
parameters = [
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[21.0, 22.0, 23.0, 24.0],
[0.0, 1.0, 2.0, 3.0],
]
samples = MockNestSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[1.0, 2.0, 3.0, 10.0, 5.0],
log_prior_list=[0.0, 0.0, 0.0, 0.0, 0.0],
weight_list=[1.0, 1.0, 1.0, 1.0, 1.0],
),
total_samples=10,
log_evidence=0.0,
number_live_points=5,
)
samples_range = samples.samples_within_parameter_range(
parameter_index=0, parameter_range=[-1.0, 100.0])
assert len(samples_range.parameter_lists) == 5
assert samples.parameter_lists[0] == samples_range.parameter_lists[0]
samples_range = samples.samples_within_parameter_range(
parameter_index=0, parameter_range=[1.0, 100.0])
assert len(samples_range.parameter_lists) == 1
assert samples_range.parameter_lists[0] == [21.0, 22.0, 23.0, 24.0]
samples_range = samples.samples_within_parameter_range(
parameter_index=2, parameter_range=[1.5, 2.5])
assert len(samples_range.parameter_lists) == 4
assert samples_range.parameter_lists[0] == [0.0, 1.0, 2.0, 3.0]
assert samples_range.parameter_lists[1] == [0.0, 1.0, 2.0, 3.0]
assert samples_range.parameter_lists[2] == [0.0, 1.0, 2.0, 3.0]
assert samples_range.parameter_lists[3] == [0.0, 1.0, 2.0, 3.0]
def test__acceptance_ratio_is_correct(self):
model = af.ModelMapper(mock_class_1=MockClassx4)
samples = MockNestSamples(
model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=5 * [[]],
log_likelihood_list=[1.0, 2.0, 3.0, 4.0, 5.0],
log_prior_list=5 * [0.0],
weight_list=5 * [0.0],
),
total_samples=10,
log_evidence=0.0,
number_live_points=5,
)
assert samples.acceptance_ratio == 0.5
def samples_via_sampler_from_model(self, model):
"""Create a `Samples` object from this non-linear search's output files on the hard-disk and model.
For Emcee, all quantities are extracted via the hdf5 backend of results.
Parameters
----------
model
The model which generates instances for different points in parameter space. This maps the points from unit
cube values to physical values via the priors.
paths : af.Paths
Manages all paths, e.g. where the search outputs are stored, the `NonLinearSearch` chains,
etc.
"""
parameters = self.backend.get_chain(flat=True).tolist()
log_priors = [
sum(model.log_priors_from_vector(vector=vector))
for vector in parameters
]
log_likelihoods = self.backend.get_log_prob(flat=True).tolist()
weights = len(log_likelihoods) * [1.0]
auto_correlation_time = self.backend.get_autocorr_time(tol=0)
total_walkers = len(self.backend.get_chain()[0, :, 0])
total_steps = len(self.backend.get_log_prob())
return EmceeSamples(
model=model,
samples=Sample.from_lists(model=model,
parameters=parameters,
log_likelihoods=log_likelihoods,
log_priors=log_priors,
weights=weights),
total_walkers=total_walkers,
total_steps=total_steps,
auto_correlation_times=auto_correlation_time,
auto_correlation_check_size=self.auto_correlation_check_size,
auto_correlation_required_length=self.
auto_correlation_required_length,
auto_correlation_change_threshold=self.
auto_correlation_change_threshold,
backend=self.backend,
time=self.timer.time,
)
def make_samples(model):
galaxy_0 = ag.Galaxy(redshift=0.5, light=ag.lp.EllSersic(centre=(0.0, 1.0)))
galaxy_1 = ag.Galaxy(redshift=1.0, light=ag.lp.EllSersic())
plane = ag.Plane(galaxies=[galaxy_0, galaxy_1])
parameters = [model.prior_count * [0.0], model.prior_count * [10.0]]
sample_list = Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[1.0, 2.0],
log_prior_list=[0.0, 0.0],
weight_list=[0.0, 1.0],
)
return mock.MockSamples(
model=model, sample_list=sample_list, max_log_likelihood_instance=plane
)
def make_samples():
model = af.ModelMapper(mock_class_1=MockClassx4)
parameters = [
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[0.0, 1.0, 2.0, 3.0],
[21.0, 22.0, 23.0, 24.0],
[0.0, 1.0, 2.0, 3.0],
]
return OptimizerSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=[1.0, 2.0, 3.0, 10.0, 5.0],
log_prior_list=[0.0, 0.0, 0.0, 0.0, 0.0],
weight_list=[1.0, 1.0, 1.0, 1.0, 1.0],
))
def test__converged__median_pdf_vector_and_instance(self):
parameters = [
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[0.9, 1.9],
[1.1, 2.1],
]
log_likelihood_list = 10 * [0.1]
weight_list = 10 * [0.1]
model = af.ModelMapper(mock_class=MockClassx2)
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
assert samples.pdf_converged is True
median_pdf_vector = samples.median_pdf_vector
assert median_pdf_vector[0] == pytest.approx(1.0, 1.0e-4)
assert median_pdf_vector[1] == pytest.approx(2.0, 1.0e-4)
median_pdf_instance = samples.median_pdf_instance
assert median_pdf_instance.mock_class.one == pytest.approx(1.0, 1e-1)
assert median_pdf_instance.mock_class.two == pytest.approx(2.0, 1e-1)
def test__unconverged_vector_at_lower_and_upper_sigma(self):
parameters = [
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.0, 2.0],
[1.1, 2.1],
[0.9, 1.9],
]
log_likelihood_list = 9 * [0.0] + [1.0]
weight_list = 9 * [0.0] + [1.0]
model = af.ModelMapper(mock_class=MockClassx2)
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
assert samples.pdf_converged is False
vector_at_sigma = samples.vector_at_sigma(sigma=1.0)
assert vector_at_sigma[0] == pytest.approx(((0.9, 1.1)), 1e-2)
assert vector_at_sigma[1] == pytest.approx(((1.9, 2.1)), 1e-2)
vector_at_sigma = samples.vector_at_sigma(sigma=3.0)
assert vector_at_sigma[0] == pytest.approx(((0.9, 1.1)), 1e-2)
assert vector_at_sigma[1] == pytest.approx(((1.9, 2.1)), 1e-2)
def from_results_internal(
cls,
results_internal,
model: AbstractPriorModel,
auto_correlation_settings: AutoCorrelationsSettings,
unconverged_sample_size: int = 100,
time: Optional[float] = None,
):
"""
The `Samples` classes in **PyAutoFit** provide an interface between the results of a `NonLinearSearch` (e.g.
as files on your hard-disk) and Python.
To create a `Samples` object after an `Zeus` model-fit the results must be converted from the
native format used by `Zeus` (which is a HDFBackend) to lists of values, the format used by the **PyAutoFit**
`Samples` objects.
This classmethod performs this conversion before creating a `ZeusSamples` object.
Parameters
----------
results_internal
The MCMC results in their native internal format from which the samples are computed.
model
Maps input vectors of unit parameter values to physical values and model instances via priors.
auto_correlations_settings
Customizes and performs auto correlation calculations performed during and after the search.
unconverged_sample_size
If the samples are for a search that is yet to convergence, a reduced set of samples are used to provide
a rough estimate of the parameters. The number of samples is set by this parameter.
time
The time taken to perform the model-fit, which is passed around `Samples` objects for outputting
information on the overall fit.
"""
parameter_lists = results_internal.get_chain(flat=True).tolist()
log_posterior_list = results_internal.get_log_prob(flat=True).tolist()
log_prior_list = model.log_prior_list_from(
parameter_lists=parameter_lists)
log_likelihood_list = [
log_posterior - log_prior for log_posterior, log_prior in zip(
log_posterior_list, log_prior_list)
]
weight_list = len(log_likelihood_list) * [1.0]
sample_list = Sample.from_lists(
model=model,
parameter_lists=parameter_lists,
log_likelihood_list=log_likelihood_list,
log_prior_list=log_prior_list,
weight_list=weight_list)
return ZeusSamples(
model=model,
sample_list=sample_list,
auto_correlation_settings=auto_correlation_settings,
unconverged_sample_size=unconverged_sample_size,
time=time,
results_internal=results_internal,
)
def test__converged__vector_and_instance_at_upper_and_lower_sigma(self):
parameters = [
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.0, 0.5],
[0.2, 0.3],
]
log_likelihood_list = list(range(10))
weight_list = 10 * [0.1]
model = af.ModelMapper(mock_class=MockClassx2)
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
assert samples.pdf_converged is True
vector_at_sigma = samples.vector_at_sigma(sigma=3.0)
assert vector_at_sigma[0] == pytest.approx((0.00242, 0.19757), 1e-1)
assert vector_at_sigma[1] == pytest.approx((0.30243, 0.49757), 1e-1)
vector_at_sigma = samples.vector_at_upper_sigma(sigma=3.0)
assert vector_at_sigma[0] == pytest.approx(0.19757, 1e-1)
assert vector_at_sigma[1] == pytest.approx(0.49757, 1e-1)
vector_at_sigma = samples.vector_at_lower_sigma(sigma=3.0)
assert vector_at_sigma[0] == pytest.approx(0.00242, 1e-1)
assert vector_at_sigma[1] == pytest.approx(0.30243, 1e-1)
vector_at_sigma = samples.vector_at_sigma(sigma=1.0)
assert vector_at_sigma[0] == pytest.approx((0.1, 0.1), 1e-1)
assert vector_at_sigma[1] == pytest.approx((0.4, 0.4), 1e-1)
instance_at_sigma = samples.instance_at_sigma(sigma=1.0)
assert instance_at_sigma.mock_class.one == pytest.approx((0.1, 0.1),
1e-1)
assert instance_at_sigma.mock_class.two == pytest.approx((0.4, 0.4),
1e-1)
instance_at_sigma = samples.instance_at_upper_sigma(sigma=3.0)
assert instance_at_sigma.mock_class.one == pytest.approx(0.19757, 1e-1)
assert instance_at_sigma.mock_class.two == pytest.approx(0.49757, 1e-1)
instance_at_sigma = samples.instance_at_lower_sigma(sigma=3.0)
assert instance_at_sigma.mock_class.one == pytest.approx(0.00242, 1e-1)
assert instance_at_sigma.mock_class.two == pytest.approx(0.30243, 1e-1)
def test__converged__errors_vector_and_instance_at_upper_and_lower_sigma(
self):
parameters = [
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.1, 0.4],
[0.0, 0.5],
[0.2, 0.3],
]
log_likelihood_list = list(range(10))
weight_list = 10 * [0.1]
model = af.ModelMapper(mock_class=MockClassx2)
samples = MockSamples(model=model,
samples=Sample.from_lists(
model=model,
parameter_lists=parameters,
log_likelihood_list=log_likelihood_list,
log_prior_list=10 * [0.0],
weight_list=weight_list,
))
assert samples.pdf_converged is True
errors = samples.error_magnitude_vector_at_sigma(sigma=3.0)
assert errors == pytest.approx([0.19514, 0.19514], 1e-1)
errors = samples.error_vector_at_upper_sigma(sigma=3.0)
assert errors == pytest.approx([0.09757, 0.09757], 1e-1)
errors = samples.error_vector_at_lower_sigma(sigma=3.0)
assert errors == pytest.approx([0.09757, 0.09757], 1e-1)
errors = samples.error_vector_at_sigma(sigma=3.0)
assert errors[0] == pytest.approx((0.09757, 0.09757), 1e-1)
assert errors[1] == pytest.approx((0.09757, 0.09757), 1e-1)
errors = samples.error_magnitude_vector_at_sigma(sigma=1.0)
assert errors == pytest.approx([0.0, 0.0], 1e-1)
errors_instance = samples.error_instance_at_sigma(sigma=1.0)
assert errors_instance.mock_class.one == pytest.approx(0.0, 1e-1)
assert errors_instance.mock_class.two == pytest.approx(0.0, 1e-1)
errors_instance = samples.error_instance_at_upper_sigma(sigma=3.0)
assert errors_instance.mock_class.one == pytest.approx(0.09757, 1e-1)
assert errors_instance.mock_class.two == pytest.approx(0.09757, 1e-1)
errors_instance = samples.error_instance_at_lower_sigma(sigma=3.0)
assert errors_instance.mock_class.one == pytest.approx(0.09757, 1e-1)
assert errors_instance.mock_class.two == pytest.approx(0.09757, 1e-1)
