This is an example to train a task with PPO algorithm.

Here it runs MemorizeDigits-v0 environment with 1000 iterations.

ppo_memorize_digits(ctxt=None, seed=1, batch_size=4000, max_episode_length=100)

Train PPO on MemorizeDigits-v0 environment.

  • ctxt (garage.experiment.ExperimentContext) – The experiment configuration used by Trainer to create the snapshotter.

  • seed (int) – Used to seed the random number generator to produce determinism.

  • batch_size (int) – Number of timesteps to use in each training step.

  • max_episode_length (int) – Max number of timesteps in an episode.