ray.rllib.models.tf.tf_modelv2.TFModelV2.metrics

TFModelV2.metrics() → Dict[str, numpy.array | jnp.ndarray | tf.Tensor | torch.Tensor]

Override to return custom metrics from your model.

The stats will be reported as part of the learner stats, i.e., info.learner.[policy_id, e.g. “default_policy”].model.key1=metric1

Returns:

The custom metrics for this model.