Description
Hi, thanks for the amazing project!
I might be misunderstanding something, so I wanted to confirm with you: is the project still under active development?
I tried running bash train_ppo.sh, but it seems that the script doesn't include multi-round interaction with the environment — at least not in a way I expected for training an RL-based agent that interacts with an environment. For example, after launching the environment, I couldn’t find code that handles the interaction loop with the agent.
Is this part still being developed, or am I missing something? Thanks in advance!
Additional Information
No response
Description
Hi, thanks for the amazing project!
I might be misunderstanding something, so I wanted to confirm with you: is the project still under active development?
I tried running bash
train_ppo.sh, but it seems that the script doesn't include multi-round interaction with the environment — at least not in a way I expected for training an RL-based agent that interacts with an environment. For example, after launching the environment, I couldn’t find code that handles the interaction loop with the agent.Is this part still being developed, or am I missing something? Thanks in advance!
Additional Information
No response