Olivia Cheng

Hi all, I have a fix ready that adds the suggested error message. Should I go ahead and submit a PR?

I initially got the same error as @narekmalk . After modifying the code to the below: ``` import torch from torch.utils._python_dispatch import TorchDispatchMode from torch._dynamo.testing import EagerAndRecordGraphs backend = EagerAndRecordGraphs()...

Have you tried using a different interpolation method? I tried your code usinig `interpolation=tvf.InterpolationMethod.BILINEAR` and got improved results. float32: bfloat16:

Hey @cduong-a , 2 questions for you about this: - From what I can tell, when CVAT exports back out the annotations after frames are removed, it will remove those...

I can also replicate this with a fake model that generates one result to trigger `loadRes()`. ``` class PredictiveModel(torch.nn.Module): def __init__(self, base_model): super().__init__() self.base_model = base_model def forward(self, images, targets=None):...