RLQWOP directory contains a custom implementation of the proximal policy optimization algorithm. Running python run-gym.py will train a model using one of openai's implementations. Since we have implemented the environment as a gym environment, we can use existing implementations of many popular RL algorithms to train an agent. multi-frame-qwop-v0 uses three sequential frames as the state. The default environment ( qwop-v0) encodes the state of the game as the position and angle of each of the runner's limbs.įrame-qwop-v0 encodes the state as the pixel data of the current frame of the game. gym-qwop/ to install the gym environment. There are three versions of the environment: qwop-v0, frame-qwop-v0, and multi-frame-qwop-v0. ![]() gym-qwop folder has python classes modeling QWOP as an open-ai gym environment.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |