RLlib - Multiagent new api - rllib-multi-agent-env-v0 already in registry

[zoetsekas]

RLlib - Multiagent new api - rllib-multi-agent-env-v0 already in registry

[zoetsekas]

when evaluation is disabled it proceeds further but new error appears
File "/opt/project/trading/training/model/rl/multi_agent/ppo/equity/trainer.py", line 129, in start_training_equity result = algo.train() File "/usr/local/lib/python3.10/site-packages/ray/tune/trainable/trainable.py", line 331, in train raise skipped from exception_cause(skipped) File "/usr/local/lib/python3.10/site-packages/ray/tune/trainable/trainable.py", line 328, in train result = self.step() File "/usr/local/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 873, in step train_results, train_iter_ctx = self._run_one_training_iteration() File "/usr/local/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 3156, in _run_one_training_iteration results = self.training_step() File "/usr/local/lib/python3.10/site-packages/ray/rllib/algorithms/ppo/ppo.py", line 424, in training_step return self._training_step_new_api_stack() File "/usr/local/lib/python3.10/site-packages/ray/rllib/algorithms/ppo/ppo.py", line 445, in _training_step_new_api_stack episodes, env_runner_results = synchronous_parallel_sample( File "/usr/local/lib/python3.10/site-packages/ray/rllib/execution/rollout_ops.py", line 94, in synchronous_parallel_sample stats_dicts = [worker_set.local_worker().get_metrics()] File "/usr/local/lib/python3.10/site-packages/ray/rllib/env/multi_agent_env_runner.py", line 615, in get_metrics return self.metrics.reduce() File "/usr/local/lib/python3.10/site-packages/ray/rllib/utils/metrics/metrics_logger.py", line 750, in reduce self.stats[sub_key] = stat.reduce() File "/usr/local/lib/python3.10/site-packages/ray/rllib/utils/metrics/stats.py", line 274, in reduce self.values = self._reduced_values()[1] File "/usr/local/lib/python3.10/site-packages/ray/rllib/utils/metrics/stats.py", line 539, in _reduced_values reduced = reduce_meth(values) File "/usr/local/lib/python3.10/site-packages/numpy/lib/nanfunctions.py", line 1052, in nanmean warnings.warn("Mean of empty slice", RuntimeWarning, stacklevel=2)

[simonsays1980]

@zoetsekas Thanks for filing the issue. These are both warnings that are due to some default names getting overwritten and initial metrics being NaN. These can be safely ignored.