Upload policy_config.py with huggingface_hub
Browse files- policy_config.py +5 -2
policy_config.py
CHANGED
@@ -8,17 +8,20 @@ exp_config = {
|
|
8 |
'epoch_per_collect': 10,
|
9 |
'batch_size': 320,
|
10 |
'learning_rate': 0.0003,
|
|
|
11 |
'weight_decay': 0,
|
12 |
'value_weight': 0.5,
|
13 |
'entropy_weight': 0.01,
|
14 |
'clip_ratio': 0.2,
|
15 |
'adv_norm': True,
|
16 |
-
'value_norm': '
|
17 |
'ppo_param_init': True,
|
18 |
'grad_norm': 0.5,
|
19 |
'n_sample': 3200,
|
20 |
'unroll_len': 1,
|
21 |
'deterministic_eval': True,
|
22 |
'model': {},
|
23 |
-
'cfg_type': 'PPOFPolicyDict'
|
|
|
|
|
24 |
}
|
|
|
8 |
'epoch_per_collect': 10,
|
9 |
'batch_size': 320,
|
10 |
'learning_rate': 0.0003,
|
11 |
+
'lr_scheduler': None,
|
12 |
'weight_decay': 0,
|
13 |
'value_weight': 0.5,
|
14 |
'entropy_weight': 0.01,
|
15 |
'clip_ratio': 0.2,
|
16 |
'adv_norm': True,
|
17 |
+
'value_norm': 'baseline',
|
18 |
'ppo_param_init': True,
|
19 |
'grad_norm': 0.5,
|
20 |
'n_sample': 3200,
|
21 |
'unroll_len': 1,
|
22 |
'deterministic_eval': True,
|
23 |
'model': {},
|
24 |
+
'cfg_type': 'PPOFPolicyDict',
|
25 |
+
'env_id': 'Hopper-v3',
|
26 |
+
'exp_name': 'Hopper-v3-PPO'
|
27 |
}
|