search

yandexdataschool / AgentNet

Deep Reinforcement Learning library for humans
http://agentnet.rtfd.org/
Other
301 stars 71 forks source link

deprecate preprocess_observation #90

Closed justheuristic closed 7 years ago

justheuristic commented 7 years ago

Currently we support custom function that preprocesses observations. However, openai gym has evolved to do that on it's own via gym.wrappers Seems a good time to deprecate the function

justheuristic commented 7 years ago

(long done)