# Error creating RLPredictor using restored checkpoint **URL:** \ **Category:** RLlib\ **Created:** [December 9, 2022, 12:40am UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605 "2022-12-09T00:40:08Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![steff](https://avatars.discourse-cdn.com/v4/letter/s/e36b37/32.png) [@steff](https://discuss.ray.io/u/steff)\ **Post date:** [December 9, 2022, 12:40am UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/1 "2022-12-09T00:40:08Z")
**How severe does this issue affect your experience of using Ray?** - High: It blocks me to complete my task. Hi experts. Get the following error when restoring checkpoint for a DQN model using Ray 2.1.0: Traceback (most recent call last): File “/home/stefan/PycharmProjects/RLProjects/rl\_offline\_trainer/inference\_app.py”, line 61, in predictor = RLPredictor.from\_checkpoint(checkpoint) File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/train/rl/rl\_predictor.py”, line 63, in from\_checkpoint policy = checkpoint.get\_policy(env) File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/train/rl/rl\_checkpoint.py”, line 42, in get\_policy return Policy.from\_checkpoint(checkpoint=self)[“default\_policy”] File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/rllib/policy/policy.py”, line 256, in from\_checkpoint policy\_state = pickle.load(f) File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/\_private/serialization.py”, line 89, in \_actor\_handle\_deserializer return ray.actor.ActorHandle.\_deserialization\_helper(serialized\_obj, outer\_id) File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/actor.py”, line 1281, in \_deserialization\_helper return worker.core\_worker.deserialize\_and\_register\_actor\_handle( File “python/ray/\_raylet.pyx”, line 2137, in ray.\_raylet.CoreWorker.deserialize\_and\_register\_actor\_handle File “python/ray/\_raylet.pyx”, line 2106, in ray.\_raylet.CoreWorker.make\_actor\_handle File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/\_private/function\_manager.py”, line 522, in load\_actor\_class actor\_class = self.\_load\_actor\_class\_from\_gcs( File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/\_private/function\_manager.py”, line 617, in \_load\_actor\_class\_from\_gcs class\_name = ensure\_str(class\_name) File “/home/stefan/anaconda3/envs/py38\_ray2.1/lib/python3.8/site-packages/ray/\_private/utils.py”, line 289, in ensure\_str assert isinstance(s, bytes) AssertionError Use Ray Tune to execute several trials with option to save checkpoint at end using the following: ``` # create tuner tuner = Tuner( # trainer trainer, # create tune configuration tune_config=self.create_tune_config( search_algo=search_algo, scheduler=scheduler ), # hyper-parameters param_space=self.create_param_space(), # specify run configuration run_config=RunConfig( stop=dict(training_iteration=2), checkpoint_config=CheckpointConfig(checkpoint_at_end=True), verbose=3 ) ) # run trials result_grid = tuner.fit() ``` Recreate best checkpoint and use it to create RLPredictor at which point the above error occurs: ``` # recreate checkpoint checkpoint = Checkpoint.from_directory(path=checkpoint_path) # create RLPredictor from checkpoint - errors occurs when this executes predictor = RLPredictor.from_checkpoint(checkpoint) ``` From what I can tell the checkpoint folder contains all necessary artifacts. What am I doing wrong? Thanks. Stefan ---
**Author:** ![arturn](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/arturn/32/2096_2.png) [@arturn](https://discuss.ray.io/u/arturn)\ **Post date:** [December 15, 2022, 9:04am UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/2 "2022-12-15T09:04:08Z")
Hi @steff, The code looks good. Could you turn this into a GH issue with a complete repro script? Cheers ---
**Author:** ![steff](https://avatars.discourse-cdn.com/v4/letter/s/e36b37/32.png) [@steff](https://discuss.ray.io/u/steff)\ **Post date:** [December 15, 2022, 9:57pm UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/3 "2022-12-15T21:57:30Z")
Sure. Is there a web page that describes the steps? ---
**Author:** ![arturn](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/arturn/32/2096_2.png) [@arturn](https://discuss.ray.io/u/arturn)\ **Post date:** [December 16, 2022, 8:53am UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/4 "2022-12-16T08:53:16Z")
Hi @steff , Nothing special, I can write the steps down: - Go to official ray repo - Click issues, create issue - Fill out form, include repro script and probably description of what you expected to happen vs was is happing - Post link here for reference ---
**Author:** ![jwlarocque](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/jwlarocque/32/4122_2.png) [@jwlarocque](https://discuss.ray.io/u/jwlarocque)\ **Post date:** [March 28, 2023, 6:20pm UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/5 "2023-03-28T18:20:43Z")
Hey @steff, I have the same problem in 2.3.1, did you end up creating an issue for this or finding a resolution? ---
**Author:** ![steff](https://avatars.discourse-cdn.com/v4/letter/s/e36b37/32.png) [@steff](https://discuss.ray.io/u/steff)\ **Post date:** [April 2, 2023, 3:58pm UTC](https://discuss.ray.io/t/error-creating-rlpredictor-using-restored-checkpoint/8605/6 "2023-04-02T15:58:32Z")
Created an issue for this. Here is the link: [Cannot create RLPredictor using restored checkpoint in different Ray session · Issue #33995 · ray-project/ray · GitHub](https://github.com/ray-project/ray/issues/33995)