RuntimeError: Input and output sizes should be greater than 0, but got input (W: 0) and output (W: 128)

[MaxTeselkin]

Hi! I am trying to run model inference on custom data, I was able to successfully create category space with corresponding .pth file, but when I tried running demo.py on my image I got the following error:

File "/workspaces/devit/my_demo.py", line 235, in main
    output = model(batched_inputs)[0]
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/devit/detectron2/modeling/meta_arch/devit.py", line 1168, in forward
    other_classes = interpolate(other_classes, self.T, mode='linear') # (Nxclasses) x spatial x T
  File "/devit/detectron2/modeling/meta_arch/devit.py", line 86, in interpolate
    return F.interpolate(seq, T, mode=mode) 
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/functional.py", line 3945, in interpolate
    return torch._C._nn.upsample_linear1d(input, output_size, align_corners, scale_factors)
RuntimeError: Input and output sizes should be greater than 0, but got input (W: 0) and output (W: 128)

Can it be because of the fact that I have only 1 class?

[mlzxy]

Yes. If you only have one class, consider setting the other class to all zeros instead of interpolating them.

[MaxTeselkin]

Thanks for help, I added "blank" class to category space using the following code:

if category_space is not None:
        category_space = torch.load(category_space)
        if len(category_space["label_names"]) < 2:
            category_space["label_names"].append("_blank_")
            real_prototypes = category_space["prototypes"]
            blank_prototypes = torch.zeros(1, 1024)
            category_space["prototypes"] = torch.cat((real_prototypes, blank_prototypes), 0)
        model.label_names = category_space["label_names"]
        model.test_class_weight = category_space["prototypes"].to(device)

and it worked