0

I am new to deep reinforcement learning! I am following this code for my adaptation problem (doing actions)

https://github.com/jaromiru/AI-blog/blob/master/CartPole-DQN.py

I am wondering how I can evaluate the training, I already got the average rewards, but how can I get the average Q values, loss per episode, and average absolute error. to evaluate my agent please!

I will be grateful if you can help me!

imen kanzali
  • 1
  • Do you mean how to extract the Q value from tensorflow operations as let's say a numpy array? – Constantinos Apr 11 '21 at 01:08
  • Actualy yes but not like array I want the plot of Q value per episode like here https://jaromiru.com/2016/10/21/lets-make-a-dqn-full-dqn/ – imen kanzali Apr 12 '21 at 03:56
  • https://github.com/jaromiru/AI-blog/blob/348628b105058d876001ca758b6ba59fb1726614/CartPole-DQN.py#L119 The self.brain.predictOne(s) is your Q value function. Is your question more of a Tensorflow question or you had difficulty tracking down the Q function? – Constantinos Apr 12 '21 at 05:51
  • Thank you for your response! Yes actually my difficulty is how to track down my q values – imen kanzali Apr 12 '21 at 12:10

0 Answers0