load_model: false
load_encoder: true
train_encoder: false
transfer_path: "results/hard-s0/3DBall"
transfer_path: "results/hard-s0/3DBallHard"
network_settings:
normalize: true
hidden_units: 128
transfer_path: "results/hard-s1/3DBall"
transfer_path: "results/hard-s1/3DBallHard"
transfer_path: "results/hard-s2/3DBall"
transfer_path: "results/hard-s2/3DBallHard"
transfer_path: "results/hard-s3/3DBall"
transfer_path: "results/hard-s3/3DBallHard"
transfer_path: "results/hard-s4/3DBall"
transfer_path: "results/hard-s4/3DBallHard"
reward_signal_steps_per_update: 10.0
encoder_layers: 2
policy_layers: 2
forward_layers: 2
forward_layers: 0
feature_size: 128
action_feature_size: 64
feature_size: 256
action_feature_size: 128
separate_policy_train: true
separate_policy_net: true
separate_model_train: true
train_model: false
load_action: true
train_action: false
transfer_path: "results/reacher-qr-s0/Reacher"
transfer_path: "results/sacmod_reacher-qr/Reacher"