Missed TensorFlow Dev Summit? Check out the video playlist. Watch recordings

tf_agents.trajectories.time_step.TimeStep

View source on GitHub

Returned with every call to step and reset on an environment.

@staticmethod
tf_agents.trajectories.time_step.TimeStep(
    _cls, step_type, reward, discount, observation
)

A TimeStep contains the data emitted by an environment at each step of interaction. A TimeStep holds a step_type, an observation (typically a NumPy array or a dict or list of arrays), and an associated reward and discount.

The first TimeStep in a sequence will equal StepType.FIRST. The final TimeStep will equal StepType.LAST. All other TimeSteps in a sequence will equal `StepType.MID.

Attributes:

  • step_type: a Tensor or array of StepType enum values.
  • reward: a Tensor or array of reward values.
  • discount: A discount value in the range [0, 1].
  • observation: A NumPy array, or a nested dict, list or tuple of arrays.

Methods

is_first

View source

is_first()

is_last

View source

is_last()

is_mid

View source

is_mid()