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

1.32k stars 131 forks source link

Open abhisheksushil2003 opened 10 months ago

abhisheksushil2003 commented 10 months 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():