search

facebookresearch / Pearl

A Production-ready Reinforcement Learning AI Agent Library brought by the Applied Reinforcement Learning team at Meta.
MIT License
2.44k stars 144 forks source link

PPO on continuous actions #77

Open zaksemenov opened 3 months ago

zaksemenov commented 3 months ago

I noticed that in the PPO agent initialization it forces the is_action_continuous=False whereas the PPO algorithm and other libraries implementing PPO allow continuous actions. Can this be added to Pearl as well

https://github.com/facebookresearch/Pearl/blob/main/pearl/policy_learners/sequential_decision_making/ppo.py#L99

yiwan-rl commented 3 months ago

Thanks for your suggestion. We actually plan to add continuous control PPO soon.