Getting Started
Usage Guide (How-To)
Tutorials
Algorithms and Methods
Reference Guide
Development Guide
API Reference
garage
garage.envs
garage.experiment
garage.np
garage.plotter
garage.replay_buffer
garage.sampler
garage.tf
garage.torch
multi_env_trpo
This is an example to train multiple tasks with TRPO algorithm.
Train TRPO on two different PointEnv instances.
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.