PushBlock:
trainer_type: ppo
hyperparameters:
batch_size: 128
buffer_size: 2048
batch_size: 1024
buffer_size: 10240
lambd: 1.0
lambd: 0.95
num_layers: 3
num_layers: 2
vis_encode_type: simple
reward_signals:
extrinsic:
time_horizon: 64
summary_freq: 10000
threaded: true
env_settings:
num_envs: 3