diff --git a/ding/bonus/config.py b/ding/bonus/config.py index c449c3fdc5..285eff6586 100644 --- a/ding/bonus/config.py +++ b/ding/bonus/config.py @@ -21,6 +21,9 @@ def get_instance_config(env_id: str, algorithm: str) -> EasyDict: cfg.learning_rate = 1e-3 cfg.action_space = 'continuous' cfg.n_sample = 1024 + elif env_id == 'Pendulum-v1': + cfg.action_space = 'continuous' + cfg.n_sample = 400 elif env_id == 'acrobot': cfg.learning_rate = 1e-4 cfg.n_sample = 400 diff --git a/ding/bonus/ppof.py b/ding/bonus/ppof.py index 149b42c3c1..bf6012240f 100644 --- a/ding/bonus/ppof.py +++ b/ding/bonus/ppof.py @@ -26,6 +26,7 @@ class PPOF: 'LunarLander-v2', 'LunarLanderContinuous-v2', 'BipedalWalker-v3', + 'Pendulum-v1', 'acrobot', # ch2: action 'rocket_landing',