ray/rllib/tuned_examples/regression_tests/cartpole-a2c-microbatch.yaml

11 lines
247 B
YAML

cartpole-a2c-microbatch:
env: CartPole-v0
run: A2C
stop:
episode_reward_mean: 100
timesteps_total: 100000
config:
num_workers: 1
gamma: 0.95
microbatch_size: 50
train_batch_size: 100