ray/rllib/tuned_examples/sac/cartpole-continuous-pybullet-sac.yaml

17 lines
420 B
YAML

cartpole-sac:
env: CartPoleContinuousBulletEnv-v0
run: SAC
stop:
episode_reward_mean: 80
timesteps_total: 100000
config:
# Works for both torch and tf.
framework: tf
gamma: 0.95
no_done_at_end: false
horizon: 200
n_step: 3
prioritized_replay: true
initial_alpha: 0.1
learning_starts: 256
clip_actions: false