lambd: 0.95
num_epoch: 3
learning_rate_schedule: constant
encoder_layers: 3
encoder_layers: 2
policy_layers: 0
policy_layers: 1
value_layers: 2
value_layers: 1
feature_size: 128
action_feature_size: 128
reuse_encoder: true
predict_return: true
use_bisim: false
separate_value_train: true
in_batch_alter: true
network_settings:
normalize: true
hidden_units: 512
load_value: false
transfer_path: "results/cs-tr-s1/CrawlerStatic"
use_transfer: true