Open tfzee opened 4 years ago
It did not converge for me and it was very slow. So i did some changes it also improves performance it solves cartpole after 30 episodes
Changes to hyperparameters actual train in batches in the replay method(far better perfomance on big batches) stop episode when done
It did not converge for me and it was very slow. So i did some changes it also improves performance it solves cartpole after 30 episodes
Changes to hyperparameters actual train in batches in the replay method(far better perfomance on big batches) stop episode when done