starry-sky6688/MARL-Algorithms

Issues

• 您好，您的代码写的非常好，请问怎么看您发表的论文呢，我想拜读您的论文，这样会更加受益匪浅

  #120 opened a month ago by lpplpf
  1

• 您好，VDN和qmix为什么不能进行多智能体通信呢

  #118 opened 3 months ago by lpplpf
  2

• 关于get_action_weights的问题

  #117 opened 5 months ago by honey-mxy
  1

• 关于在别的环境应用qmix出现episodes rewards dropout的问题

  #115 opened 6 months ago by ayton-zhang
  2

• 关于训练得到的模型的问题

  #116 opened 6 months ago by honey-mxy
  1

• 关于attention的训练依据的问题

  #114 opened 6 months ago by honey-mxy
  1

• 关于evaluate的胜率

  #113 opened 6 months ago by honey-mxy
  0

• 关于GA-Common和GA-AC的问题

  #112 opened 6 months ago by issta24
  1

• Quick Start 会报错，请问是什么问题。

  #111 opened 7 months ago by 1patrickBob
  2

• Translate code comments to English

  #110 opened 8 months ago by gabrielgcbs
  0

• 关于qtran的问题

  #109 opened 8 months ago by Johnson221b
  1

• 关于qtran_base.py中_get_individual_q的一个小问题

  #108 opened 9 months ago by Johnson221b
  2

• 策略函数中的eval_hidden和target_hidden如何理解

  #106 opened 9 months ago by Johnson221b
  2

• None

  #107 opened 9 months ago by Johnson221b
  0

• custom data traing

  #105 opened 10 months ago by shandongchong
  1

• 自定义的环境能使用这里面的算法跑吗？

  #104 opened a year ago by shswdmx1
  1

• 关于参数reuse_network

  #100 opened 2 years ago by zanezhenli
  3

• 可以使用其他的环境跑这里面的算法吗？

  #103 opened a year ago by DurableFrank
  1

• 关于g2anet中hard_weights的问题

  #102 opened 2 years ago by Icbyone
  1

• 关于COMA critic网络输入

  #101 opened 2 years ago by hjh0119
  3

• 关于QMIX的Trick：Eligibility traces

  #99 opened 2 years ago by zanezhenli
  1

• 关于涉及环境参数的一些疑问

  #98 opened 2 years ago by zanezhenli
  1

• 请问avail_action是什么呢？

  #97 opened 2 years ago by Shuang-ZS
  1

• 请问CommNet 和 G2ANet 需要外部训练算法是什么意思呢

  #96 opened 2 years ago by Shuang-ZS
  1

• 智能体决策不同时

  #95 opened 2 years ago by Clown1206
  1

• A bug when choosing actions

  #94 opened 2 years ago by Chty-syq
  1

• mavan的cuda有问题....

  #93 opened 2 years ago by harnvo
  1

• 内存泄漏问题

  #90 opened 2 years ago by hccz95
  4

• 为什么改名哇

  #89 opened 2 years ago by zichuan-liu
  1

• QMIX 目标网络更新

  #88 opened 3 years ago by L-Spike
  1

• 为何没有保存model?

  #87 opened 3 years ago by lbqdcxy
  4

• 訓練時間

  #86 opened 3 years ago by OPlobo
  1

• 关于reinforce算法中的问题

  #85 opened 3 years ago by Duke-Allen
  2

• 咨询一个关于qmix的问题

  #84 opened 3 years ago by zhouweiqing-star
  1

• Its version is StarCraft 2 version: SC2 4.10? and How do I cite this project

  #83 opened 3 years ago by zichuan-liu
  1

• 可视化界面

  #82 opened 3 years ago by dameng123
  7

• Question about get action weights

  #81 opened 3 years ago by LXXXXR
  1

• 关于网络更新时的问题

  #80 opened 3 years ago by Duke-Allen
  3

• def forward(self, q_values, states): # states的shape为(episode_num, max_episode_len， state_shape) # 传入的q_values是三维的，shape为(episode_num, max_episode_len， n_agents)

  #79 opened 3 years ago by zhouweiqing-star
  1

• Comments in English?

  #78 opened 3 years ago by balamir53
  1

• 策略梯度算法Reinforce的问题

  #77 opened 3 years ago by Duke-Allen
  4

• 引入新的算法

  #76 opened 3 years ago by Bingoyww
  1

• 关于loss计算的问题

  #75 opened 3 years ago by Duke-Allen
  1

• 关于连续动作的问题

  #74 opened 3 years ago by Kevinstone-199898
  2

• 关于胜率图像的问题

  #73 opened 3 years ago by Duke-Allen
  1

• Should probably detach these values right?

  #72 opened 3 years ago by aravindvenu7
  1

• 关于restore的问题

  #71 opened 3 years ago by Kevinstone-199898
  2

• 关于reinforce算法的问题

  #70 opened 3 years ago by Duke-Allen
  3

• 关于设置batch大小

  #69 opened 3 years ago by Duke-Allen
  4

• 函数get_obs_agent()中数字表示意义

  #68 opened 3 years ago by JiaoAlvin
  0