Is it possible to use DiCENet train a regression model

[w11m]

thanks for the great work i would like to use the idea of DiceNet on a custom dataset. And i'm new to DL and pytorch can you give me some idea how to make it?

Thank you:)

[w11m]

And also my data is a non-3 channel (70 channel)

[w11m]

I edited the program for own dataloader and try to change all the arguments(channel,width,height) in the code

but still get the error message below

/home/tan/William/DiCENeT/EdgeNets/nn_layers/dice.py:73: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!
  if height != self.height:
/home/tan/William/DiCENeT/EdgeNets/nn_layers/dice.py:83: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!
  if height != h_wise_height:
/home/tan/William/DiCENeT/EdgeNets/nn_layers/dice.py:91: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!
  if width != self.width:
/home/tan/William/DiCENeT/EdgeNets/nn_layers/dice.py:100: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!
  if width != w_wise_width:
Traceback (most recent call last):
  File "train_classification.py", line 346, in <module>
    main(args)
  File "train_classification.py", line 83, in main
    flops = compute_flops(model)
  File "/home/tan/William/DiCENeT/EdgeNets/utilities/utils.py", line 64, in compute_flops
    _ = model(input)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/tan/William/DiCENeT/EdgeNets/model/classification/dicenet.py", line 169, in forward
    x = self.level1(x)  # 112
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/tan/William/DiCENeT/EdgeNets/nn_layers/cnn_utils.py", line 54, in forward
    return self.cbr(x)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/container.py", line 92, in forward
    input = module(input)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/conv.py", line 343, in forward
    return self.conv2d_forward(input, self.weight)
  File "/usr/local/lib/python3.5/dist-packages/torch/nn/modules/conv.py", line 340, in conv2d_forward
    self.padding, self.dilation, self.groups)
**RuntimeError: Given groups=1, weight of size 16 70 3 3, expected input[1, 3, 224, 224] to have 70 channels, but got 3 channels instead**

the data size is 96 x 96 x 70 (W x H x C)

[w11m]

I think I already fix the error above the error caused by the code in /utilities/utils.py

input = input if input is not None else torch.Tensor(1, 70, 96, 96)

def compute_flops(model, input=None):
    from utilities.flops_compute import add_flops_counting_methods
    input = input if input is not None else torch.Tensor(1, 70, 96, 96)

    model = add_flops_counting_methods(model)
    model.eval()
    model.start_flops_count()

    _ = model(input)

    flops = model.compute_average_flops_cost()  # + (model.classifier.in_features * model.classifier.out_features)
    flops = flops / 1e6 / 2
    return flops