gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-18 03:30:19 +01:00
Code Issues Projects Releases Wiki Activity

bug fix - preventing crashes when the probability of one of the actions is 0 in the policy head

Browse Source
This commit is contained in:
Itai Caspi
2017-10-31 10:49:50 +02:00
parent 1918f16079
commit 913ab75e8a
4 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
agents/actor_critic_agent.py
View File

@@ -121,7 +121,7 @@ class ActorCriticAgent(PolicyOptimizationAgent):
else:
action = np.argmax(action_probabilities)
action_info = {"action_probability": action_probabilities[action], "state_value": state_value}
self.entropy.add_sample(-np.sum(action_probabilities * np.log(action_probabilities)))
self.entropy.add_sample(-np.sum(action_probabilities * np.log(action_probabilities + eps)))
else:
# CONTINUOUS
state_value, action_values_mean, action_values_std = self.main_network.online_network.predict(observation)

2
agents/policy_gradients_agent.py
View File

@@ -73,7 +73,7 @@ class PolicyGradientsAgent(PolicyOptimizationAgent):
else:
action = np.argmax(action_values)
action_value = {"action_probability": action_values[action]}
self.entropy.add_sample(-np.sum(action_values * np.log(action_values)))
self.entropy.add_sample(-np.sum(action_values * np.log(action_values + eps)))
else:
# CONTINUOUS
result = self.main_network.online_network.predict(observation)