ray/rllib/tuned_examples/td3/pendulum-td3-fake-gpus.yaml

23 lines
592 B
YAML

pendulum-td3-fake-gpus:
env: Pendulum-v1
run: TD3
stop:
episode_reward_mean: -900
timesteps_total: 100000
config:
# Works for both torch and tf.
framework: tf
actor_hiddens: [64, 64]
critic_hiddens: [64, 64]
replay_buffer_config:
type: MultiAgentReplayBuffer
num_steps_sampled_before_learning_starts: 5000
exploration_config:
random_timesteps: 5000
evaluation_interval: 10
evaluation_duration: 5
# Fake 2 GPUs.
num_gpus: 2
_fake_gpus: true