RicardoDominguez / PyCREPS

Contextual Relative Entropy Policy Search for Reinforcement Learning in Python
14 stars 1 forks source link

CartPole solved condition #12

Closed RicardoDominguez closed 5 years ago

RicardoDominguez commented 5 years ago

OpenAI gym considers the cartPole problem solved when the reward over 100 episodes is >= 195 link.

Currently considered solved when the reward over 100 episodes is >= 200.

RicardoDominguez commented 5 years ago

Update documentation.