MaloneLin's repositories
gym
A toolkit for developing and comparing reinforcement learning algorithms.
Language:PythonNOASSERTION000
lunarlanderDDPG
fine turn version of DDPG for LunarLanderContinuous-v2
Language:PythonMIT000
Policy-Gradient-Lunarlander
The best result of LunarLander-v2 using deep reinforcement learning algorithm Policy Gradient reinforce I could ever try. By testing 200 times. Test reward>=200:(189/200). Avg steps:285. Avg reward: 248.78
Language:PythonMIT000