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

10 lines
215 B
YAML

cartpole-es:
env: CartPole-v0
run: ES
stop:
episode_reward_mean: 75
timesteps_total: 400000
config:
num_workers: 2
noise_size: 25000000
episodes_per_batch: 50