reinforcement-learning
reinforcement-learning copied to clipboard
Reset BlackjackEnv to a chosen state
I want to run multiple MC trajectories from the same start state. How do I reset the BlackjackEnv to a certain start state? Thanks.
BlackjackEnv
does not have reset(self, observation)
, which would be suitable. You'd need to extend BlackjackEnv
with this function. Make a pull request to openai-gym.