I got this user warning when I tried to implement the augmentation for my data, has anyone met this problem?
UserWarning: Argument fill/fillcolor is not supported for Tensor input. Fill value is zero
warnings.warn(“Argument fill/fillcolor is not supported for Tensor input. Fill value is zero”)
You’re probably using torchvision v0.8.2 or older. This issue was fixed 5 months ago in the PR #2904. If you are not using v0.9.0 or newer, you won’t be able to use fill in a Tensor input
So, the only solution is to upgrade your torchvision.