functionality for random and regular complete
cartpole experiments
This commit is contained in:
parent
b902e07424
commit
62f0e15881
@ -222,7 +222,7 @@ class ActiveRLService(Node):
|
||||
|
||||
self.active_rl_pub.publish(rl_response)
|
||||
self.end_time = time.time()
|
||||
self.get_logger().info(f'RL Time: {self.end_time-self.begin_time}, mode: {self.interactive_run}')
|
||||
# self.get_logger().info(f'RL Time: {self.end_time-self.begin_time}, mode: {self.interactive_run}')
|
||||
self.begin_time = None
|
||||
self.end_time = None
|
||||
|
||||
@ -263,7 +263,7 @@ class ActiveRLService(Node):
|
||||
self.rl_pending = False
|
||||
|
||||
self.end_time = time.time()
|
||||
self.get_logger().info(f'RL Time: {self.end_time - self.begin_time}, mode: {self.interactive_run}')
|
||||
# self.get_logger().info(f'RL Time: {self.end_time - self.begin_time}, mode: {self.interactive_run}')
|
||||
self.begin_time = None
|
||||
self.end_time = None
|
||||
|
||||
@ -278,7 +278,7 @@ class ActiveRLService(Node):
|
||||
self.active_rl_pub.publish(rl_response)
|
||||
|
||||
self.end_time = time.time()
|
||||
self.get_logger().info(f'RL Time: {self.end_time - self.begin_time}, mode: {self.interactive_run}')
|
||||
# self.get_logger().info(f'RL Time: {self.end_time - self.begin_time}, mode: {self.interactive_run}')
|
||||
|
||||
self.begin_time = None
|
||||
self.end_time = None
|
||||
|
@ -123,6 +123,7 @@ class ActiveBOTopic(Node):
|
||||
self.seed_array = None
|
||||
self.env = None
|
||||
self.active_bo_pending = False
|
||||
self.BO = None
|
||||
|
||||
def active_bo_callback(self, msg):
|
||||
if not self.active_bo_pending:
|
||||
@ -408,7 +409,7 @@ class ActiveBOTopic(Node):
|
||||
|
||||
state_msg = ActiveBOState()
|
||||
state_msg.current_run = self.current_run + 1 if self.current_run < self.bo_runs else self.bo_runs
|
||||
state_msg.current_episode = self.current_episode + 1 \
|
||||
state_msg.current_episode = self.current_episode \
|
||||
if self.current_episode < self.bo_episodes else self.bo_episodes
|
||||
state_msg.best_reward = float(self.best_reward)
|
||||
state_msg.last_user_reward = self.last_user_reward
|
||||
|
Loading…
Reference in New Issue
Block a user