Hello, I am new to Ray and attempting to run some basic examples, with an end goal of using IMPALA to run distributed policy optimization across multiple cluster GPU nodes.
I have followed the default installation here RLlib: Industry-Grade Reinforcement Learning — Ray 1.12.0 but have not been able to find any working simple examples thus far. Is there a better source of code examples or getting started guide for Ray RLlib or IMPALA usage in particular?
Ex 1
source : url above (html link count per post limited)
error 1:
$HOME/.conda/envs/rllib/lib/python3.8/site-packages/ray/rllib/agents/trainer.py", line 1035, in _init
raise NotImplementedError
error 2:
$HOME/.conda/envs/rllib/lib/python3.8/site-packages/ray/_private/services.py", line 392, in wait_for_node
raise TimeoutError(“Timed out while waiting for node to startup.”)
Ex 2
source : ray/impala.py at master · ray-project/ray · GitHub
error :
Traceback (most recent call last):
File “cartpole.py”, line 1, in
from ray.rllib.agents.impala import ImpalaConfig
ImportError: cannot import name ‘ImpalaConfig’ from ‘ray.rllib.agents.impala’
($HOME/.conda/envs/rllib/lib/python3.8/site-packages/ray/rllib/agents/impala/init.py)
Thanks for any guidance or recommendations!
How severe does this issue affect your experience of using Ray?
- Medium: It contributes to significant difficulty to complete my task, but I can work around it.