def samples(self):
"""Returns the samples in the chain as a FieldArray.
If the sampling args are not the same as the model params, the
returned samples will have both the sampling and the model params.
The returned FieldArray has dimension [additional dimensions x]
nwalkers x niterations.
"""
# chain is a [additional dimensions x] niterations x ndim array
samples = self.chain
sampling_params = self.sampling_params
# convert to dictionary to apply boundary conditions
samples = {
param: samples[..., ii]
for ii, param in enumerate(sampling_params)
}
samples = self.model.prior_distribution.apply_boundary_conditions(
**samples)
# now convert to field array
samples = FieldArray.from_arrays(
[samples[param] for param in sampling_params],
names=sampling_params)
# apply transforms to go to model params space
if self.model.sampling_transforms is not None:
samples = self.model.sampling_transforms.apply(samples,
inverse=True)
return samples
def prior_rvs(self, size=1, prior=None):
"""Returns random variates drawn from the prior.
If the ``sampling_params`` are different from the ``variable_params``,
the variates are transformed to the `sampling_params` parameter space
before being returned.
Parameters
----------
size : int, optional
Number of random values to return for each parameter. Default is 1.
prior : JointDistribution, optional
Use the given prior to draw values rather than the saved prior.
Returns
-------
FieldArray
A field array of the random values.
"""
# draw values from the prior
if prior is None:
prior = self.prior_distribution
p0 = prior.rvs(size=size)
# transform if necessary
if self.sampling_transforms is not None:
ptrans = self.sampling_transforms.apply(p0)
# pull out the sampling args
p0 = FieldArray.from_arrays(
[ptrans[arg] for arg in self.sampling_params],
names=self.sampling_params)
return p0
def prior_rvs(self, size=1, prior=None):
"""Returns random variates drawn from the prior.
If the ``sampling_args`` are different from the ``variable_args``, the
variates are transformed to the `sampling_args` parameter space before
being returned.
Parameters
----------
size : int, optional
Number of random values to return for each parameter. Default is 1.
prior : JointDistribution, optional
Use the given prior to draw values rather than the saved prior.
Returns
-------
FieldArray
A field array of the random values.
"""
# draw values from the prior
if prior is None:
prior = self._prior
p0 = prior.rvs(size=size)
# transform if necessary
if self._sampling_transforms is not None:
ptrans = self.apply_sampling_transforms(p0)
# pull out the sampling args
p0 = FieldArray.from_arrays([ptrans[arg]
for arg in self._sampling_args],
names=self._sampling_args)
return p0
def samples(self):
"""Returns the samples in the chain as a FieldArray.
If the sampling args are not the same as the variable args, the
returned samples will have both the sampling and the variable args.
The returned FieldArray has dimension [additional dimensions x]
nwalkers x niterations.
"""
# chain is a [additional dimensions x] niterations x ndim array
samples = self.chain
sampling_args = self.sampling_args
# convert to dictionary to apply boundary conditions
samples = {
param: samples[..., ii]
for ii, param in enumerate(sampling_args)
}
samples = self.likelihood_evaluator._prior.apply_boundary_conditions(
**samples)
# now convert to field array
samples = FieldArray.from_arrays(
[samples[param] for param in sampling_args], names=sampling_args)
# apply transforms to go to variable args space
return self.likelihood_evaluator.apply_sampling_transforms(
samples, inverse=True)
def samples(self):
"""Returns the samples in the chain as a FieldArray.
If the sampling args are not the same as the variable args, the
returned samples will have both the sampling and the variable args.
The returned FieldArray has dimension [additional dimensions x]
nwalkers x niterations.
"""
# chain is a [additional dimensions x] niterations x ndim array
samples = self.chain
sampling_args = self.sampling_args
# convert to dictionary to apply boundary conditions
samples = {param: samples[...,ii]
for ii,param in enumerate(sampling_args)}
samples = self.likelihood_evaluator._prior.apply_boundary_conditions(
**samples)
# now convert to field array
samples = FieldArray.from_arrays([samples[param]
for param in sampling_args],
names=sampling_args)
# apply transforms to go to variable args space
return self.likelihood_evaluator.apply_sampling_transforms(samples,
inverse=True)
