diff --git a/README.md b/README.md index 533bac8..71e5119 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ Implemented algorithms: * Synchronous Advantage Actor Critic (A2C) * Synchronous N-Step Q-Learning * Deep Deterministic Policy Gradient (DDPG) -* Synchronous Proximal Policy Optimization (PPO) +* (Continuous/Discrete) Synchronous Proximal Policy Optimization (PPO) * Action Conditional Video Prediction Asynchronous algorithms below are removed in this repo but can be found in [the previous release](https://github.com/ShangtongZhang/DeepRL/releases/tag/v0.1) @@ -42,6 +42,7 @@ Asynchronous algorithms below are removed in this repo but can be found in [the ## PPO ![Loading...](https://raw.githubusercontent.com/ShangtongZhang/DeepRL/master/images/ppo_continuous-180408-002056.png) +![Loading...](https://raw.githubusercontent.com/ShangtongZhang/DeepRL/master/images/ppo_pixel_atari-180410-235529.png) ## Action Conditional Video Prediction ![Loading...](https://raw.githubusercontent.com/ShangtongZhang/DeepRL/master/images/ACVP.png) diff --git a/images/ppo_pixel_atari-180410-235529.png b/images/ppo_pixel_atari-180410-235529.png new file mode 100644 index 0000000..50f09b9 Binary files /dev/null and b/images/ppo_pixel_atari-180410-235529.png differ diff --git a/main.py b/main.py index c1497ee..0d927dd 100644 --- a/main.py +++ b/main.py @@ -338,11 +338,12 @@ def plot(): # 'dqn_pixel_atari-180407-01414', # 'quantile_regression_dqn_pixel_atari-180407-01604', # 'n_step_dqn_pixel_atari-180408-001104', - 'ppo_continuous-180408-002056', - 'ddpg_continuous-180407-234141' + # 'ppo_continuous-180408-002056', + # 'ddpg_continuous-180407-234141' + 'ppo_pixel_atari-180410-235529', ] for name in names: - plotter.plot_results(['to_plot/%s' % (name)]) + plotter.plot_results(['to_plot/%s' % (name)], title='BreakoutNoFrameskip-v4') plt.savefig('images/%s.png' % (name)) plt.close()