wassname/pytorch-soft-actor-critic
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-soft-actor-critic.git synced 2026-08-12 12:20:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update main.py

Browse Source
This commit is contained in:
Pranjal Tandon
2018-09-12 09:39:49 +05:30
committed by GitHub
parent 3d2b410d0c
commit 28d86af462
1 changed files with 0 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main.py
-21
View File
@@ -85,26 +85,5 @@ for i_episode in itertools.count():
plot_line(total_numsteps, rewards, args)
print("Episode: {}, total numsteps: {}, reward: {}, average reward: {}".format(i_episode, total_numsteps, np.round(rewards[-1],2),
np.round(np.mean(rewards[-100:]),2)))
if i_episode % 50 == 0 and args.eval==True:
state = torch.Tensor([env.reset()])
episode_reward = 0
while True:
action = agent.select_action(state, deterministic=True)
next_state, reward, done, _ = env.step(action)
episode_reward += reward
next_state = torch.Tensor([next_state])
state = next_state
if done:
break
rewards_test.append(episode_reward)
print("-----------------------------------Test-----------------------------------------------")
print("Episode: {}, total numsteps: {}, reward: {}, average reward: {}".format(i_episode, total_numsteps,
np.round(rewards_test[-1],2),
np.round(np.mean(rewards_test[-10:]),2)))
print("--------------------------------------------------------------------------------------")
env.close()