|
|
|
|
|
|
|
learning_rate: 0.0003 |
|
|
|
learning_rate_schedule: constant |
|
|
|
batch_size: 256 |
|
|
|
buffer_size: 500000 |
|
|
|
buffer_size: 1000000 |
|
|
|
buffer_init_steps: 2000 |
|
|
|
tau: 0.005 |
|
|
|
steps_per_update: 20.0 |
|
|
|
|
|
|
|
encoder_layers: 2 |
|
|
|
policy_layers: 1 |
|
|
|
forward_layers: 2 |
|
|
|
policy_layers: 0 |
|
|
|
forward_layers: 0 |
|
|
|
# separate_policy_train: true |
|
|
|
separate_value_net: true |
|
|
|
reuse_encoder: false |
|
|
|
separate_policy_train: true |
|
|
|
# separate_model_train: true |
|
|
|
# separate_value_net: true |
|
|
|
reuse_encoder: true |
|
|
|
in_epoch_alter: false |
|
|
|
in_batch_alter: true |
|
|
|
use_op_buffer: false |
|
|
|
|
|
|
|
use_transfer: true |
|
|
|
load_model: true |
|
|
|
train_model: false |
|
|
|
transfer_path: "results/cs-sacmod-ov-old/CrawlerStatic" |
|
|
|
transfer_path: "results/cs-sacmod-linear/CrawlerStatic" |
|
|
|
network_settings: |
|
|
|
normalize: true |
|
|
|
hidden_units: 512 |
|
|
|
|
yanchaosun