With one exception, your dataset cannot be subjected to those transforms: anticipate scalar type but discovered to float Double #Exception: It's not possible to apply those transforms to your dataset: expected scalar type Float but found Double

microsoft / computervision-recipes

Best Practices, code samples, and documentation for Computer Vision.

MIT License

9.47k stars 1.17k forks source link

With one exception, your dataset cannot be subjected to those transforms: anticipate scalar type but discovered to float Double #Exception: It's not possible to apply those transforms to your dataset: expected scalar type Float but found Double #676

Open thisishusseinali opened 2 years ago

thisishusseinali commented 2 years ago

Description 👍🏼

Thank you for this fantastic project. I attempt to execute 02 state of the art.ipynb, but encounter an error that contains the address; my machine is running Cuda10.1, and my GPU is a Rtx2080ti. I've upgraded the versions of PyTorch and TorchVision to 1.7.0 and 0.8.0, respectively.