Skip to content

cagpjax.policies.orthogonalization

Classes:

OrthogonalizationPolicy 

OrthogonalizationPolicy(base_policy: AbstractBatchLinearSolverPolicy, method: OrthogonalizationMethod = OrthogonalizationMethod.QR, n_reortho: int = 0)

Bases: AbstractBatchLinearSolverPolicy

Orthogonalization policy.

This policy orthogonalizes (if necessary) the action operator produced by the base policy.

Parameters:

  • base_policy

    (AbstractBatchLinearSolverPolicy) –

    The base policy that produces the action operator to be orthogonalized.

  • method

    (OrthogonalizationMethod, default: QR ) –

    The method to use for orthogonalization.

  • n_reortho

    (int, default: 0 ) –

    The number of times to re-orthogonalize each column. Reorthogonalizing once is generally sufficient to improve orthogonality for Gram-Schmidt variants (see e.g. 10.1007/s00211-005-0615-4).

Source code in src/cagpjax/policies/orthogonalization.py 
27
28
29
30
31
32
33
34
35
def __init__(
    self,
    base_policy: AbstractBatchLinearSolverPolicy,
    method: OrthogonalizationMethod = OrthogonalizationMethod.QR,
    n_reortho: int = 0,
):
    self.base_policy = base_policy
    self.method = method
    self.n_reortho = n_reortho