MixedPolicy

class itergp.methods.policies.MixedPolicy(base_policies, iters)

Bases: LinearSolverPolicy

Mixed policy.

Policy which chooses actions based on a set of base policies.

Parameters

• base_policies (Tuple[policies.LinearSolverPolicy]) – Policies which make up the MixedPolicy.

• iters (Tuple[int]) – Until which iteration (non-inclusive) to use the policy in the corresponding position in base_policies. Assumed to be sorted in increasing order. If iters has one fewer entry than base_policies, the last policy is used for all remaining iterations.

Return type

None

Methods Summary

__call__(solver_state[, rng])

Return an action for a given solver state.

Methods Documentation

__call__(solver_state, rng=None)

Return an action for a given solver state.

Parameters

• solver_state (probnum.linalg.solvers.LinearSolverState) – Current state of the linear solver.

• rng (Optional[SeedSequence]) – Random number generator.

Returns

Next action to take.

Return type

action