From 2ac3759688a9928619d5f10410aafcd214042f46 Mon Sep 17 00:00:00 2001
From: "Egli Adrian (IT-SCI-API-PFI)" <adrian.egli@sbb.ch>
Date: Wed, 6 Jan 2021 16:05:45 +0100
Subject: [PATCH] Tensorboard support added
---
reinforcement_learning/rl_agent_test.py | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/reinforcement_learning/rl_agent_test.py b/reinforcement_learning/rl_agent_test.py
index 5295971..8687563 100644
--- a/reinforcement_learning/rl_agent_test.py
+++ b/reinforcement_learning/rl_agent_test.py
@@ -3,6 +3,7 @@ from collections import namedtuple
import gym
import numpy as np
+from torch.utils.tensorboard import SummaryWriter
from reinforcement_learning.dddqn_policy import DDDQNPolicy
from reinforcement_learning.ppo_agent import PPOPolicy
@@ -36,6 +37,9 @@ def cartpole(use_dddqn=False):
episode = 0
checkpoint_interval = 20
scores_window = deque(maxlen=100)
+
+ writer = SummaryWriter()
+
while True:
episode += 1
state = env.reset()
@@ -79,6 +83,10 @@ def cartpole(use_dddqn=False):
policy.memory)),
end=" ")
+ writer.add_scalar("CartPole/value", tot_reward, episode)
+ writer.add_scalar("CartPole/smoothed_value", np.mean(scores_window), episode)
+ writer.flush()
+
if __name__ == "__main__":
cartpole()
--
GitLab