tfp.experimental.mcmc.step_kernel

Takes num_steps repeated TransitionKernel steps from current_state.

tfp.experimental.mcmc.step_kernel(
    num_steps, current_state, previous_kernel_results=None, kernel=None,
    return_final_kernel_results=False, parallel_iterations=10, seed=None, name=None
)

This is meant to be a minimal driver for executing TransitionKernels; for something more featureful, see sample_chain.

num_steps Integer number of Markov chain steps.
current_state Tensor or Python list of Tensors representing the current state(s) of the Markov chain(s).
previous_kernel_results A Tensor or a nested collection of Tensors. Warm-start for the auxiliary state needed by the given kernel. If not supplied, step_kernel will cold-start with kernel.bootstrap_results.
kernel An instance of tfp.mcmc.TransitionKernel which implements one step of the Markov chain.
return_final_kernel_results If True, then the final kernel results are returned alongside the chain state after num_steps steps are taken. This can be useful to inspect the final auxiliary state, or for a later warm restart.
parallel_iterations The number of iterations allowed to run in parallel. It must be a positive integer. See tf.while_loop for more details.
seed Optional, a seed for reproducible sampling.
name Python str name prefixed to Ops created by this function. Default value: None (i.e., 'mcmc_step_kernel').

next_state Markov chain state after num_step steps are taken, of identical type as current_state.
final_kernel_results kernel results, as supplied by kernel.one_step after num_step steps are taken. This is only returned if return_final_kernel_results is True.