How severe does this issue affect your experience of using Ray? High: It blocks me to complete my task. Hi all, I am trying to load in a previously trained model to continue training it, except I get the following error: Failure # 1 (occurred at 2023-03-31_14-54-08) e[36mray::PPO.train()e[39m …

RuntimeError: Expected scalars to be on CPU, got cuda:0 instead

Jules_Damji May 23, 2023, 5:37pm 7

@Denys_Ashikhin May be close this issue since it’s fixed in Ray 2.5, and we are in the process of releasing it in the coming days?

1 Like

Topic		Replies	Views
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_addmm) RLlib	4	2880	August 8, 2022
Error when trying to use gpus during RL training RLlib	4	645	July 21, 2021
Questions about using GPU for the ray[rllib] RLlib	4	1973	August 4, 2023
Error occuring repeatedly during training at random time steps RLlib	3	50	January 3, 2025
Running out of CUDA Memory - Sample Batch , rollout fragment length is not helping RLlib	2	942	July 28, 2022