gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-17 11:10:20 +01:00
Code Issues Projects Releases Wiki Activity

imporved API for getting / setting variables within the graph

Browse Source
This commit is contained in:
Itai Caspi
2017-10-25 16:07:58 +03:00
parent e33b0e8534
commit 1918f16079
5 changed files with 45 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
agents/clipped_ppo_agent.py
View File

@@ -16,7 +16,6 @@
from agents.actor_critic_agent import *
from random import shuffle
import tensorflow as tf
# Clipped Proximal Policy Optimization - https://arxiv.org/abs/1707.06347
@@ -132,7 +131,7 @@ class ClippedPPOAgent(ActorCriticAgent):
loss[key] = np.mean(loss[key], 0)
if self.tp.learning_rate_decay_rate != 0:
curr_learning_rate = self.tp.sess.run(self.tp.learning_rate)
curr_learning_rate = self.main_network.online_network.get_variable_value(self.tp.learning_rate)
self.curr_learning_rate.add_sample(curr_learning_rate)
else:
curr_learning_rate = self.tp.learning_rate