save_replay_buffer: false
init_entcoef: 1.0
reward_signal_steps_per_update: 20.0
encoder_layers: 2
policy_layers: 1
forward_layers: 1
encoder_layers: 3
policy_layers: 0
forward_layers: 0
feature_size: 128
feature_size: 300
action_layers: 2
action_feature_size: 128
separate_policy_train: true
train_model: false
load_action: true
train_action: false
transfer_path: "results/oldcs/CrawlerStatic"
transfer_path: "results/oldcs-f300/CrawlerStatic"
network_settings:
normalize: true
hidden_units: 512