ray/rllib/tuned_examples/regression_tests/cartpole-ppo-tf.yaml
2020-01-20 23:06:50 -08:00

10 lines
241 B
YAML

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