tf_agents.bandits.policies.linalg.conjugate_gradient

Returns x such that A * x = b.

tf_agents.bandits.policies.linalg.conjugate_gradient(
    a_mat: tf_agents.typing.types.Tensor,
    b: tf_agents.typing.types.Tensor,
    tol: float = 1e-10
) -> tf_agents.typing.types.Float

Implements the Conjugate Gradient method. https://en.wikipedia.org/wiki/Conjugate_gradient_method

a_mat a Symmetric Positive Definite matrix, represented as a Tensor of shape [n, n].
b a Tensor of shape [n, 1].
tol (float) desired tolerance on the residual.

x Tensor x of shape [n, 1] such that A * x = b.

ValueError if a_mat is not square or a_mat and b have incompatible shapes.