2020-05-26 11:10:27 +02:00
|
|
|
halfcheetah-ppo:
|
|
|
|
|
env: HalfCheetah-v2
|
|
|
|
|
run: PPO
|
|
|
|
|
stop:
|
|
|
|
|
episode_reward_mean: 9800
|
|
|
|
|
time_total_s: 10800
|
|
|
|
|
config:
|
2020-05-27 16:19:13 +02:00
|
|
|
# Works for both torch and tf.
|
|
|
|
|
framework: tf
|
2020-05-26 11:10:27 +02:00
|
|
|
gamma: 0.99
|
|
|
|
|
lambda: 0.95
|
|
|
|
|
kl_coeff: 1.0
|
|
|
|
|
num_sgd_iter: 32
|
|
|
|
|
lr: .0003
|
|
|
|
|
vf_loss_coeff: 0.5
|
|
|
|
|
clip_param: 0.2
|
|
|
|
|
sgd_minibatch_size: 4096
|
|
|
|
|
train_batch_size: 65536
|
|
|
|
|
num_workers: 16
|
|
|
|
|
num_gpus: 1
|
|
|
|
|
grad_clip: 0.5
|
|
|
|
|
num_envs_per_worker:
|
|
|
|
|
grid_search: [16, 32]
|
|
|
|
|
batch_mode: truncate_episodes
|
|
|
|
|
observation_filter: MeanStdFilter
|
