def _build_mcmc_init_fn(
self,
prior: Any,
potential_fn: Callable,
init_strategy: str = "prior",
**kwargs,
) -> Callable:
"""
Return function that, when called, creates an initial parameter set for MCMC.
Args:
prior: Prior distribution.
potential_fn: Potential function that the candidate samples are weighted
with.
init_strategy: Specifies the initialization method. Either of
[`prior`|`sir`|`latest_sample`].
kwargs: Passed on to init function. This way, init specific keywords can
be set through `mcmc_parameters`. Unused arguments should be absorbed.
Returns: Initialization function.
"""
if init_strategy == "prior":
return lambda: prior_init(prior, **kwargs)
elif init_strategy == "sir":
return lambda: sir(prior, potential_fn, **kwargs)
elif init_strategy == "latest_sample":
latest_sample = IterateParameters(self._mcmc_init_params, **kwargs)
return latest_sample
else:
raise NotImplementedError
def _build_mcmc_init_fn(
self,
prior: Any,
potential_fn: Callable,
init_strategy: str = "prior",
init_strategy_num_candidates: int = 10000,
**kwargs,
) -> Callable:
"""
Return function that, when called, creates an initial parameter set for MCMC.
Args:
prior: Prior distribution.
potential_fn: Potential function that the candidate samples are weighted
with.
init_strategy: Specifies the initialization method. Either of
[`prior`|`sir`].
init_strategy_num_candidates: Number of candidate samples drawn.
kwargs: Absorbs passed but unused arguments. E.g. in
`DirectPosterior.sample()` we pass `mcmc_parameters` which might
contain entries that are not used here.
Returns: Initialization function.
"""
if init_strategy == "prior":
return lambda: prior_init(prior)
elif init_strategy == "sir":
return lambda: sir(
prior,
potential_fn,
init_strategy_num_candidates,
)
elif init_strategy == "latest_sample":
return lambda: self._mcmc_init_params
else:
raise NotImplementedError