ray/rllib/tuned_examples/regression_tests/cartpole-appo-tf.yaml

15 lines
328 B
YAML
Raw Normal View History

cartpole-appo-tf:
env: CartPole-v0
run: APPO
stop:
episode_reward_mean: 150
timesteps_total: 200000
config:
use_pytorch: false
rollout_fragment_length: 10
train_batch_size: 10
num_envs_per_worker: 5
num_workers: 1
num_gpus: 0
vtrace: false