# Env not recognized when used with Tuner.restore **URL:** \ **Category:** Checkpointing, Restoring\ **Created:** [November 27, 2023, 4:39pm UTC](https://discuss.ray.io/t/env-not-recognized-when-used-with-tuner-restore/12955 "2023-11-27T16:39:30Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![dylan906](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/dylan906/32/3556_2.png) [@dylan906](https://discuss.ray.io/u/dylan906)\ **Post date:** [November 27, 2023, 4:39pm UTC](https://discuss.ray.io/t/env-not-recognized-when-used-with-tuner-restore/12955/1 "2023-11-27T16:39:30Z")
When I try to restore an incomplete tune job using this script: ```python """Resume experiment script.""" # %% Imports import os import ray from ray.rllib.models import ModelCatalog from ray.tune import Tuner from ray.tune.registry import register_env from punchclock.nets.lstm_mask import MaskedLSTM from punchclock.ray.build_env import buildEnv # %% Register model and Env ModelCatalog.register_custom_model("MaskedLSTM", MaskedLSTM) register_env("ssa_env", buildEnv) checkpoint_dir = "/home/user/ray_results/exp_name" num_cpus = 20 num_workers = num_cpus - 1 ray.init(num_cpus=num_cpus, num_gpus=0) os.environ["TUNE_MAX_PENDING_TRIALS_PG"] = str(num_workers) tuner = Tuner.restore( trainable="PPO", path=checkpoint_dir, resume_errored=True, restart_errored=True, ) tuner.fit() ``` I get the following error, which indicates that the environment is not recognized.But in the above script, I register the environment using `ray.tune.registry.register_env`, so I don’t know why this error would occur. There is also a second failture that I am unsure is related to the environment error. ```auto Failure # 1 (occurred at 2023-11-20_20-39-26) The actor died because of an error raised in its creation task, e[36mray::PPO. __init__ ()e[39m (pid=130077, ip=10.128.8.91, actor_id=1969b2ae33b28f9c1b1fa87701000000, repr=PPO) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/worker_set.py", line 242, in _setup self.add_workers( File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/worker_set.py", line 635, in add_workers raise result.get() File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/utils/actor_manager.py", line 488, in __fetch_result result = ray.get(r) ray.exceptions.RayActorError: The actor died because of an error raised in its creation task, e[36mray::RolloutWorker. __init__ ()e[39m (pid=131032, ip=10.128.8.91, actor_id=73949f41d65591a253439f3e01000000, repr=) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 569, in make _check_version_exists(ns, name, version) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 219, in _check_version_exists _check_name_exists(ns, name) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 197, in _check_name_exists raise error.NameNotFound( gymnasium.error.NameNotFound: Environment ssa_env doesn't exist. During handling of the above exception, another exception occurred: e[36mray::RolloutWorker. __init__ ()e[39m (pid=131032, ip=10.128.8.91, actor_id=73949f41d65591a253439f3e01000000, repr=) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/rollout_worker.py", line 609, in __init__ self.env = env_creator(copy.deepcopy(self.env_context)) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/env/utils.py", line 178, in _gym_env_creator raise EnvError(ERR_MSG_INVALID_ENV_DESCRIPTOR.format(env_descriptor)) ray.rllib.utils.error.EnvError: The env string you provided ('ssa_env') is: a) Not a supported/installed environment. b) Not a tune-registered environment creator. c) Not a valid env class string. Try one of the following: a) For Atari support: `pip install gym[atari] autorom[accept-rom-license]`. For VizDoom support: Install VizDoom (https://github.com/mwydmuch/ViZDoom/blob/master/doc/Building.md) and `pip install vizdoomgym`. For PyBullet support: `pip install pybullet`. b) To register your custom env, do `from ray import tune; tune.register('[name]', lambda cfg: [return env obj from here using cfg])`. Then in your config, do `config['env'] = [name]`. c) Make sure you provide a fully qualified classpath, e.g.: `ray.rllib.examples.env.repeat_after_me_env.RepeatAfterMeEnv` During handling of the above exception, another exception occurred: e[36mray::PPO. __init__ ()e[39m (pid=130077, ip=10.128.8.91, actor_id=1969b2ae33b28f9c1b1fa87701000000, repr=PPO) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 475, in __init__ super(). __init__ ( File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/tune/trainable/trainable.py", line 170, in __init__ self.setup(copy.deepcopy(self.config)) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 601, in setup self.workers = WorkerSet( File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/worker_set.py", line 194, in __init__ raise e.args[0].args[2] ray.rllib.utils.error.EnvError: The env string you provided ('ssa_env') is: a) Not a supported/installed environment. b) Not a tune-registered environment creator. c) Not a valid env class string. Try one of the following: a) For Atari support: `pip install gym[atari] autorom[accept-rom-license]`. For VizDoom support: Install VizDoom (https://github.com/mwydmuch/ViZDoom/blob/master/doc/Building.md) and `pip install vizdoomgym`. For PyBullet support: `pip install pybullet`. b) To register your custom env, do `from ray import tune; tune.register('[name]', lambda cfg: [return env obj from here using cfg])`. Then in your config, do `config['env'] = [name]`. c) Make sure you provide a fully qualified classpath, e.g.: `ray.rllib.examples.env.repeat_after_me_env.RepeatAfterMeEnv` Failure # 2 (occurred at 2023-11-25_19-33-43) e[36mray::PPO.train()e[39m (pid=217206, ip=10.128.8.112, actor_id=67791a70455479e668d7bad601000000, repr=PPO) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/tune/trainable/trainable.py", line 389, in train raise skipped from exception_cause(skipped) File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/tune/trainable/trainable.py", line 386, in train result = self.step() File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 832, in step results = self._compile_iteration_results( File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/algorithms/algorithm.py", line 3046, in _compile_iteration_results results["sampler_results"] = summarize_episodes( File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/metrics.py", line 221, in summarize_episodes filt = [v for v in v_list if not np.any(np.isnan(v))] File "/home/user/.conda/envs/punch/lib/python3.10/site-packages/ray/rllib/evaluation/metrics.py", line 221, in filt = [v for v in v_list if not np.any(np.isnan(v))] TypeError: ufunc 'isnan' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe'' ``` The custom environment works normally if I am initializing a new tuning run; I only see this problem when attempting to restore an experiment from a checkpoint. If anyone has any ideas on where to look, I’d much appreciate it.