Test_shape failed with torchsparse 1.2

[YoushaaMurhij]

Good work! I manged to get the shape using Test_shape.py but when using test_shape.py, I faced this error:

root@PC:/home/adelai_ws/AdelaiDepth# python3 LeReS/tools/test_shape.py --load_ckpt res101.pth --backbone resnext101
No protocol specified
processing (0000)-th image... LeReS/test_images/5.jpg
Traceback (most recent call last):
  File "LeReS/tools/test_shape.py", line 123, in <module>
    shift, focal_length, depth_scaleinv = reconstruct3D_from_depth(rgb, pred_depth_ori,
  File "LeReS/tools/test_shape.py", line 76, in reconstruct3D_from_depth
    shift_1 = refine_shift(pred_depth_norm, shift_model, predicted_focal_1, cam_u0, cam_v0)
  File "/home/adelai_ws/AdelaiDepth/LeReS/lib/test_utils.py", line 124, in refine_shift
    shift = refine_shift_one_step(depth_wshift_tmp, model, focal, u0, v0)
  File "/home/adelai_ws/AdelaiDepth/LeReS/lib/test_utils.py", line 108, in refine_shift_one_step
    outputs = model(inputs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/trainer/adelai_ws/AdelaiDepth/LeReS/lib/spvcnn_classsification.py", line 148, in forward
    x2 = self.stage2(x1)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/container.py", line 141, in forward
    input = module(input)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/trainer/adelai_ws/AdelaiDepth/LeReS/lib/spvcnn_classsification.py", line 23, in forward
    out = self.net(x)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/container.py", line 141, in forward
    input = module(input)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/torchsparse/nn/modules/conv.py", line 72, in forward
    return conv3d(inputs,
  File "/usr/local/lib/python3.8/dist-packages/torchsparse/nn/functional/conv.py", line 118, in conv3d
    idx_query = list(convert_neighbor_map_gpu(idx_query))
  File "/usr/local/lib/python3.8/dist-packages/torchsparse/nn/functional/convert_neighbor_map.py", line 9, in forward
    idx_batch, idx_point = torch.where(neighbor_map != -1)
ValueError: not enough values to unpack (expected 2, got 1)

I tried another version (1.4) but there is a change in the API. Any suggestions? Thanks!

[srikanth-midde]

Hi, Can any help me to sort out this issue? I got the same "ValueError: not enough values to unpack (expected 2, got 1)" error. Though I tried with both torchsparse v1.2 and v1.4. but facing the same issue.

[isamabdullah88]

Hey you can replace first line with second line with torchsparse==1.4 Replace this: from torchsparse.utils import sparse_collate_fn, sparse_quantize with this: from torchsparse.utils.collate import sparse_collate_fn from torchsparse.utils.quantize import sparse_quantize

It'll work with torchsparse==1.4

[isamabdullah88]

Hey, Please check for default values for models in the test_shape.py and test_depth.py scripts. Those values are different (resnet50 and resnet101), and they have to be the same.

[Ir1d]

@isamabdullah88 What do you mean by default values? the values are the same @YoushaaMurhij Can you please share how you solved this?

[isamabdullah88]

@Ir1d

• Firstly move the model files to specified location and make sure the weight files are being read by the script. Place a check in code for that. I think the above error happens because of weight files not available.
• Then, make sure that --load_ckpt and --backbone both use the same model. (Either resnet101.pth or resnet50.pth).

Hope this helps!