IWLSKernel#
- class liesel.goose.iwls.IWLSKernel(position_keys, chol_info_fn=None, initial_step_size=0.01, da_target_accept=0.8, da_gamma=0.05, da_kappa=0.75, da_t0=10)[source]#
Bases:
ModelMixin
,TransitionMixin
[IWLSKernelState
,DefaultTransitionInfo
]An IWLS kernel with dual averaging and an (optional) user-defined function for computing the Cholesky decomposition of the Fisher information matrix, implementing the
liesel.goose.types.Kernel
protocol.Methods
This section is empty if this class has only inherited attributes.
end_epoch
(prng_key, kernel_state, ...)Sets the step size as found by the dual averaging algorithm.
end_warmup
(prng_key, kernel_state, ...)Currently does nothing.
init_state
(prng_key, model_state)Initializes the kernel state.
start_epoch
(prng_key, kernel_state, ...)Resets the state of the dual averaging algorithm.
tune
(prng_key, kernel_state, model_state, epoch)Currently does nothing.
Attributes
This section is empty if this class has only inherited attributes.
Dict of error codes and their meaning.
Kernel identifier, set by
EngineBuilder
Whether this kernel needs its history for tuning.
Tuple of position keys handled by this kernel.