tf_agents.trajectories.time_step_spec

Returns a TimeStep spec given the observation_spec.

View aliases

Main aliases

tf_agents.trajectories.time_step.time_step_spec

tf_agents.trajectories.time_step_spec(
    observation_spec: Optional[types.NestedSpec] = None,
    reward_spec: Optional[types.NestedSpec] = None
) -> tf_agents.trajectories.TimeStep

Used in the notebooks

Used in the tutorials
Policies Tutorial on Multi Armed Bandits in TF-Agents A Tutorial on Multi-Armed Bandits with Per-Arm Features

Args
observation_spec	A nest of tf.TypeSpec or ArraySpec objects.
reward_spec	(Optional) A nest of tf.TypeSpec or ArraySpec objects. Default - a scalar float32 of the same type (Tensor or Array) as observation_spec.

Returns
A TimeStep with the same types (TypeSpec or ArraySpec) as the first entry in observation_spec.

Raises
TypeError	If observation and reward specs aren't both either tensor type specs or array type specs.