An example to train a task with DQN algorithm.

Here it creates a gym environment CartPole, and trains a DQN with 50k steps.

dqn_cartpole(ctxt=None, seed=1)

Train TRPO with CubeCrash-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.