UnitVectorPolicy¶
- class itergp.methods.policies.UnitVectorPolicy(ordering='lexicographic')¶
Bases:
LinearSolverPolicyStandard unit vector policy.
Policy returning standard unit vectors according to a given ordering.
- Parameters
ordering (str) – Ordering strategy of the rows (and columns) of the system matrix.
- Return type
None
Attributes Summary
Ordering strategy defining in which order to select datapoints.
Methods Summary
__call__(solver_state, rng)Return an action for a given solver state.
Attributes Documentation
- ordering¶
Ordering strategy defining in which order to select datapoints.
Methods Documentation
- __call__(solver_state, rng)[source]¶
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