User Guide¶
Run PlugRL end to end: start a server, then start one or more env clients.
Quickstart¶
plugrl-run-server fpo-policy default fpo default \
--policy.device cpu --algo.global-steps 500000 --algo.buffer-size 4096
plugrl-run-env-client mujoco-v1 --server-host 127.0.0.1 --server-port 8000 \
--num-envs 1 --num-episodes 600 --runner.replan-steps 1 --runner.seed 0
That pair learns. Get Started explains the two flags that are not optional, and has the dummy connectivity check.
Verify¶
- Server prints a WebSocket listening address.
- Env client prints server metadata and steps episodes.
Workflow¶
- Start a training server with
plugrl-run-server. (There is alsoplugrl-run-server-ray, but it is not a supported path today - see Get Started.) - Start one or more env clients with
plugrl-run-env-client <env_id>.
Components¶
plugrl-server: batches inference across connected workers, runs learning and checkpointingplugrl-env-client: creates Gymnasium envs, sendsinfer, receivesaction, sendsfeedbackplugrl-protocol: WebSocket transport, message types, and msgpack serialization
Common options¶
- Server default address is
0.0.0.0:8000. - Env client connects via
--server-hostand--server-port.
Troubleshooting¶
- Env client keeps retrying: confirm the server is listening and the address is reachable.
- Policy or algorithm not found in CLI: ensure the registration module is imported before the CLI is built.