DreamerEnv

torchrl.envs.model_based.dreamer.DreamerEnv(*args, **kwargs)

Dreamer simulation environment.

This environment is used for imagination rollouts in Dreamer training. It never terminates (done is always False) since imagination runs for a fixed horizon. The done-checking methods are overridden to avoid CUDA synchronization overhead from Python control flow on CUDA tensors.