ray.rllib.policy.eager_tf_policy_v2.EagerTFPolicyV2.extra_learn_fetches_fn#

EagerTFPolicyV2.extra_learn_fetches_fn() → Dict[str, Union[numpy.array, jnp.ndarray, tf.Tensor, torch.Tensor]][source]#

Extra stats to be reported after gradient computation.

Returns

An extra fetch-dict.

Return type

Dict[str, TensorType]