AttributeError: module 'torch._C' has no attribute '_get_privateuse1_backend_name'

MCG-NJU / VideoMAE

[NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

https://arxiv.org/abs/2203.12602

Other

1.39k stars 137 forks source link

Open abhisheksushil2003 opened 1 year ago

abhisheksushil2003 commented 1 year ago

When running the code for pretraining on kinetics data set, without distributing it across multiple nodes. I am getting the following error.

AttributeError: module 'torch._C' has no attribute '_get_privateuse1_backend_name'

from this line in the dataloader.py file. elif self._pin_memory_device == torch._C._get_privateuse1_backend_name():