Metropolis Random Walk MCMC¶
- class pints.MetropolisRandomWalkMCMC(x0, sigma0=None)[source]¶
Metropolis Random Walk MCMC, as described in [1].
Metropolis using multivariate Gaussian distribution as proposal step, also known as Metropolis Random Walk MCMC. In each iteration (t) of the algorithm, the following occurs:
propose x' ~ N(x_t, Sigma) generate u ~ U(0, 1) calculate r = pi(x') / pi(x_t) if r > u, x_t+1 = x'; otherwise, x_t+1 = x_t
here Sigma is the covariance matrix of the proposal.
Extends
SingleChainMCMC
.References
- in_initial_phase()¶
For methods that need an initial phase (see
needs_initial_phase()
), this method returnsTrue
if the method is currently configured to be in its initial phase. For other methods aNotImplementedError
is returned.
- n_hyper_parameters()¶
Returns the number of hyper-parameters for this method (see
TunableMethod
).
- needs_initial_phase()¶
Returns
True
if this method needs an initial phase, for example an adaptation-free period for adaptive covariance methods, or a warm-up phase for DREAM.
- needs_sensitivities()¶
Returns
True
if this methods needs sensitivities to be passed in totell
along with the evaluated logpdf.
- set_hyper_parameters(x)¶
Sets the hyper-parameters for the method with the given vector of values (see
TunableMethod
).- Parameters:
x – An array of length
n_hyper_parameters
used to set the hyper-parameters.
- set_initial_phase(in_initial_phase)¶
For methods that need an initial phase (see
needs_initial_phase()
), this method toggles the initial phase algorithm. For other methods aNotImplementedError
is returned.