wangzixiang
/
lerobot
mirror of https://github.com/huggingface/lerobot.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

nit 

Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
This commit is contained in:
Michel Aractingi 2025-02-13 17:12:57 +01:00
parent b07d95f0dd
commit 95de8e273d
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
lerobot/common/policies/sac/configuration_sac.py
View File

@ -95,6 +95,6 @@ class SACConfig:
"use_tanh_squash": True, "use_tanh_squash": True,
"log_std_min": -5, "log_std_min": -5,
"log_std_max": 2, "log_std_max": 2,
"init_final": 0.01, "init_final": 0.005,
} }
) )

4
lerobot/scripts/server/actor_server.py
View File

@ -163,10 +163,10 @@ def serve_actor_service(port=50052):
def update_policy_parameters(policy: SACPolicy, parameters_queue: queue.Queue, device): def update_policy_parameters(policy: SACPolicy, parameters_queue: queue.Queue, device):
if not parameters_queue.empty(): if not parameters_queue.empty():
logging.debug("[ACTOR] Load new parameters from Learner.") logging.info("[ACTOR] Load new parameters from Learner.")
state_dict = parameters_queue.get() state_dict = parameters_queue.get()
state_dict = move_state_dict_to_device(state_dict, device=device) state_dict = move_state_dict_to_device(state_dict, device=device)
policy.load_state_dict(state_dict) policy.load_state_dict(state_dict, strict=False)
def act_with_policy(cfg: DictConfig, robot: Robot, reward_classifier: nn.Module): def act_with_policy(cfg: DictConfig, robot: Robot, reward_classifier: nn.Module):