ray/rllib/tuned_examples/regression_tests/cartpole-ppo.yaml

10 lines
238 B
YAML

cartpole-ppo:
env: CartPole-v0
run: PPO
stop:
episode_reward_mean: 150
timesteps_total: 100000
config:
num_workers: 1
batch_mode: complete_episodes
observation_filter: MeanStdFilter