You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello,
I have been working on the prerequisite task on ESPNet. I am working inside a virtual environment for openvino. I cloned the model's repo and I am trying to run inference using the pytorch version of the model where i call a script written by the authors for inference, but it gives me the following error:
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.
Outside openvino environment, CUDA is detetcted while inside openvino env it is not. I installed all the requirements and followed the instructions for installation in anaconda. I tried a lot with different versions for torch.
Can someone tell me what to do?
Thanks
The text was updated successfully, but these errors were encountered:
@Nouran-Muhammad please try to load model on CPU instead. We can not use nvidia GPU in openvino environment because not all users can have this hardware, also installing torch without GPU support significantly reduce the memory required for installing package.
The solution which you need to use is already provided in your error message: If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU., please add this parameter
Hello,
I have been working on the prerequisite task on ESPNet. I am working inside a virtual environment for openvino. I cloned the model's repo and I am trying to run inference using the pytorch version of the model where i call a script written by the authors for inference, but it gives me the following error:
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.
Outside openvino environment, CUDA is detetcted while inside openvino env it is not. I installed all the requirements and followed the instructions for installation in anaconda. I tried a lot with different versions for torch.
Can someone tell me what to do?
Thanks
The text was updated successfully, but these errors were encountered: