It is assumed that you have completed reinforcement learning 13. If you put Pendulum-v0 in the jupyter notebook of reinforcement learning 13, ... Is useless. See the ChainerRL / GitHub sample. https://github.com/chainer/chainerrl/tree/master/examples/gym
train_dqn_gym.py itself gave better results. Rewrite (November 26) Fly to Reinforcement Learning 20.
Recommended Posts