MJeremy2017/reinforcement-learning-implementation

Issues

• action = self.chooseAction() sometimes returns empty

  #23 opened 4 months ago by SrikkanthProgrammer
  0

• Bug in initialising board state. Should be initialised to "dtype=np.int32"

  #20 opened 10 months ago by AbhiDu96
  0

• Q-learning code crashing after sometimes

  #16 opened 2 years ago by Adubi
  1

• Potential Error in Class: Dyna-Q+: should not reset self.time to zero

  #17 opened 2 years ago by MariosGkMeng
  0

• FileNotFoundError: [Errno 2] No such file or directory: 'policy_p1

  #15 opened 2 years ago by w1767
  1

• Show the rounds with smallest steps

  #14 opened 2 years ago by daniel-xion
  0

• Kept running non-stop when the distance between win and start states are a bit further

  #13 opened 2 years ago by daniel-xion
  1

• 'bool' object no callable

  #12 opened 2 years ago by Slurbisaur
  0

• Wrong default value for mx_nxt_reward in chooseAction

  #9 opened 4 years ago by mahsayazdani
  4

• Error concerning policies

  #2 opened 5 years ago by FalseMSP
  1

• Missing closing parenthesis causing syntax error

  #6 opened 4 years ago by maxshowarth
  1

• TD Lambda - Reset z

  #4 opened 4 years ago by thiagolcks
  1

• Showing the grid world

  #1 opened 5 years ago by maryamghoo
  0