git clone https://github.com/kbys-t/gym_PO.git
cd gym_PO
pip install -e .
-
First of all,
import gym_pomdp
-
Select environment from ``["AcrobotPO-v0", "AcrobotPO-v1", "BallArmPO-v0", "BallArmPO-v1"]`
ENV_NAME = "AcrobotPO-v0"
env = gym.make(ENV_NAME)
-
Give degree of POMDP
env.POMDP_PARAM
within[0.0, 1.0]
-
Send action and get observation and reward
observation, reward,, done, info = env.step(action)