kwai/DouZero

Issues

• 新版本在Windows下仍然无法Train?

  #27 opened 3 years ago
  1

• 使用训练模型的过程中，会出现普通炸大于王炸的情况

  #26 opened 3 years ago
  5

• SL模型的训练数据是如何实现和加载的？

  #25 opened 3 years ago
  3

• Error when fighting with RLCard in landlord_up position

  #24 opened 3 years ago
  1

• how to avoid interrupt, when learner model loads weights to the actor model while it is running 'forward' in 'act' function

  #23 opened 3 years ago
  1

• why use 'Lock' in "get_batch()" while using queue to communication between threads

  #22 opened 3 years ago
  3

• why use multiple threads to execute the function "batch_and_learn" in dmc.py.

  #21 opened 3 years ago
  2

• 为什么evaluate出错？

  #20 opened 3 years ago
  6

• Actions generate by model differ from online demo

  #19 opened 3 years ago
  1

• 关于模型版本的问题

  #18 opened 3 years ago
  1

• env/env.py _get_obs_landlord函数的疑问

  #17 opened 3 years ago
  1

• 请问基于强化学习？数据怎么得到？

  #16 opened 3 years ago
  1

• jim

  #15 opened 3 years ago
  0

• 弄

  #14 opened 3 years ago
  0

• 关于算法

  #13 opened 3 years ago
  3

• 斗得过人吗？

  #12 opened 3 years ago
  1

• 2个问题

  #10 opened 3 years ago
  3

• 666

  #9 opened 3 years ago
  0

• Ch

  #8 opened 3 years ago
  1

• 什么时候支持欢乐斗地主？

  #7 opened 3 years ago
  1

• 能协助我斗地主吗

  #6 opened 3 years ago
  18

• 这算游戏bug吗?

  #5 opened 3 years ago
  1

• 亲测，效果一般。对战水平低下，我这样一般人胜利都在50%以上

  #3 opened 3 years ago
  2

• 请问能实现像AlfaGo一样用于与人类对战吗？

  #2 opened 3 years ago
  3

• lol~ , interesting job

  #1 opened 3 years ago
  1