Learn the latest in machine learning, generative AI, and more at WiML Symposium 2023Register

tf_agents.agents.ppo.ppo_utils.make_trajectory_mask

Mask boundary trajectories and those with invalid returns and advantages.

tf_agents.agents.ppo.ppo_utils.make_trajectory_mask(
    batched_traj: tf_agents.trajectories.Trajectory
) -> tf_agents.typing.types.Tensor

Args
`batched_traj`	Trajectory, doubly-batched [batch_dim, time_dim,...]. It must be preprocessed already.

Returns
A mask, type tf.float32, that is 0.0 for all between-episode Trajectory (batched_traj.step_type is LAST) and 0.0 if the return value is unavailable.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2023-10-06 UTC.

Args