Hi,
When copying from the example in Ray RLlib, I find code like this:
> config = (
PPOConfig() # Also try common gym envs like: "CartPole-v1" or "Pendulum-v1". .environment( MultiAgentCustomRenderedEnv if args.multi_agent else CustomRenderedEnv, env_config={"corridor_length": 10, "max_steps": 100}, ) .framework(args.framework) # Use a vectorized env with 2 sub-envs. .rollouts(num_envs_per_worker=2, num_rollout_workers=1) .evaluation( # Evaluate once per training iteration. evaluation_interval=1, # Run evaluation on (at least) two episodes evaluation_duration=2, # ... using one evaluation worker (setting this to 0 will cause # evaluation to run on the local evaluation worker, blocking # training until evaluation is done). evaluation_num_workers=1, # Special evaluation config. Keys specified here will override # the same keys in the main config, but only for evaluation. evaluation_config=PPOConfig.overrides( # Render the env while evaluating. # Note that this will always only render the 1st RolloutWorker's # env and only the 1st sub-env in a vectorized env. render_env=True, ), ) )
However, it fails with:
AttributeError: type object ‘PPOConfig’ has no attribute ‘overrides’
Thanks!