Shichang Liu

> version？updated？ 2.0.0, latest version

> cannot reproduce with Python 3.8.13, change python version may solve this problem It's weird, my 3.8.13 doesn't work but 3.7 works fine

> that not the wirdest situation,i got two virtual environment ,which both used python3.8.13,but one of the environment report this error that exactly same with yours,but the other environment run...

> @Loovelj @Li-Lai @Klawens @bobo0810 请问你们Linux还是Mac？ Linux, ubuntu 18.04

I have same issue with you.Did you solve it?

same problem

It's common to scale the loss by a constant factor to control the magnitude of gradients during training. Large gradients can lead to unstable training, so scaling the loss down...

same problem

same

Problem solved, use from torch.utils.data.dataset import Dataset class TheDataset(Dataset):