你好，请问我在运行maac_run 文件的时候提示缺少这个文件，请问这个怎么解决，谢谢您

Error Message：

epoch2073
Traceback (most recent call last):
File "e: /wuzhu/Documents/Visual Studio Code/EdgeFed-MARL-MEC-master/AC_run.py", line 88, in
AC. train(MAX_EPOCH, MAX_EP_STEPS, up_freq=up_freq, render=True, render_freq=render_freq)
File "e: \wuzhu\Documents\Visual Studio Code\EdgeFed-MARL-MEC-master\AC_agent.py", line 379, in train
cur_reward = self.actor_act (epoch)
File "e: \wuzhu\Documents \Visual studio Code\EdgeFed-MARL-MEC-master \AC_agent.py", line 243, in actor_act
move _ori = circle_argmax(move _dist, self.env.move_r)
File "e: \wuzhu\Documents \isual Studio Code\EdgeFed-MARL-MEC-master\AC_agent.py", line 156, in circle_argmax
return max_pos[np.argmin(pos_dist)]
File "< array functioninternals>", line 6, in argmin
"F: \anaconda3\envs\test\lib\site-packages \numpy\core\fromnumeric.py", line 1269, in argmin
return _wrapfunc(a,'argmin', axis=axis, out=out)
File "F: \anaconda3\envs\test\lib\site-packages\numpy\core\fromnumeric.py" line 58, in _wrapfunc
return bound (*args,**kwds)
ValueError: attempt to get argmin of an empty sequence

`

Code Section：AC_agent.py

def circle_argmax(move_dist, move_r):
max_pos = np.argwhere(tf.squeeze(move_dist, axis=-1) == np.max(move_dist))
# print(tf.squeeze(move_dist, axis=-1))
pos_dist = np.linalg.norm(max_pos - np.array([move_r, move_r]), axis=1)
# print(max_pos)
return max_pos[np.argmin(pos_dist)]
`

Describe：

Hello, when I was training your code, after the 2073rd round of training, I reported the above error. Have you encountered a similar error message? How was it resolved?

Hi,
I recently read your paper <Federated Multi-Agent Actor-Critic Learning for Age Sensitive Mobile Edge Computing* [J]. IEEE Internet of Things Journal, 2021> and am interested about your model. I have tried to simulate your code but have following questions:

1. Do I need to first use 'pip -install -e' command to set up the MEC_env environment before simulate other python files?
2. When I tried to run the *_run.py files, it reports 'ImportError: cannot import name 'Space' from 'gym.spaces' error, so which version of gym did you use when simulating your code?

Thank you very much for your time if you can answer my quries.

Dear Dr. Zhu,

I came across your research on deep reinforcement learning algorithms, specifically the use of Mixed DDPG and H-MAAC in multi-agent environments. I found your work on H-MAAC to be very interesting and insightful.

I am particularly interested in your research on multi-agent deep reinforcement learning and the use of the MADDPG algorithm. I understand that your current implementation only includes the MIxed DDPG and H-MAAC algorithms and would like to inquire if you also have code for the MADDPG algorithm.

If possible, may I request access to the MADDPG code that you have developed for your research? Your code can help me deepen my understanding of the MADDPG algorithm and its applications in multi-agent MEC environments. My email address is [email protected]

Thank you for your time and consideration. I look forward to hearing from you soon.

Best regards,
Haozhe li