RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking arugment for argument index in method wrapper_index_select)

pchhapolika · May 11, 2023, 2:17pm

Hi @ehalit , I am also facing the similar issue when training on 2 or more GPU’s. Do I need to change my DemoDataset(Dataset): class? RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:1 and cuda:0!

Kindly help, please!

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 index in method wrapper__index_select) Beginners	2	1500	July 30, 2023
Resume_from_checkpoints leads to RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! 🤗Transformers	4	492	November 13, 2023
RuntimeError - NPL with transformers book - 02_classification.ipynb Beginners	1	323	April 30, 2022
Vilt RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu) in the new update Models	0	361	January 23, 2024
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.LongTensor [1, 128]] is at version 3; expected version 2 instead. Hint: the backtrace further above shows the operation that failed t 🤗Transformers	0	1380	July 20, 2023