英文

TQC代理玩RocketLander-v0

这是一个经过训练的TQC代理模型,用于玩RocketLander-v0,使用了 stable-baselines3 library RL Zoo

RL Zoo是一个训练框架,用于Stable Baselines3强化学习代理,包括超参数优化和预训练代理。

用法(使用SB3 RL Zoo)

RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo SB3: https://github.com/DLR-RM/stable-baselines3 SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib

Gym环境: https://github.com/sdsubhajitdas/Rocket_Lander_Gym

# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo tqc --env RocketLander-v0 -orga araffin -f logs/
python enjoy.py --algo tqc --env RocketLander-v0  -f logs/

训练(使用RL Zoo)

python train.py --algo tqc --env RocketLander-v0 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo tqc --env RocketLander-v0 -f logs/ -orga araffin

超参数

OrderedDict([('env_wrapper',
              [{'rl_zoo3.wrappers.FrameSkip': {'skip': 4}},
               {'rl_zoo3.wrappers.HistoryWrapper': {'horizon': 2}}]),
             ('n_timesteps', 3000000.0),
             ('policy', 'MlpPolicy'),
             ('normalize', False)])