A mistake occurred while training cyclegan

[infororange]

when i use
python train.py --dataroot ./datasets/horse2zebra --name horse2zebra --model cycle_gan
There will be errors :
NotSupportedError: slicing multiple dimensions at the same time isn't supported yet
proposals (Tensor): boxes to be encoded
"""
# perform some unpacking to make it JIT-fusion friendly
wx = weights[0]
wy = weights[1]
ww = weights[2]
wh = weights[3]

proposals_x1 = proposals[:, 0].unsqueeze(1)
               ~~~~~~~~~ <--- HERE
proposals_y1 = proposals[:, 1].unsqueeze(1)
proposals_x2 = proposals[:, 2].unsqueeze(1)
proposals_y2 = proposals[:, 3].unsqueeze(1)

reference_boxes_x1 = reference_boxes[:, 0].unsqueeze(1)
reference_boxes_y1 = reference_boxes[:, 1].unsqueeze(1)
reference_boxes_x2 = reference_boxes[:, 2].unsqueeze(1)
reference_boxes_y2 = reference_boxes[:, 3].unsqueeze(1)

How should this be solved?thank you so much!

[junyanz]

Not sure if it is relevant to our repo. We didn't use reference_boxes in this codebase.

[zhihaoWang1994]

你好 我们也遇到了这个问题 请问您解决了么

[mojoee]

Same Problem right here even when training with the standard configurations as described in the github

[junyanz]

It might be caused by the incompatibility between your PyTorch version and torchvision version.

[mojoee]

@junyanz that's right. I fixed it by uninstalling the pytorch libraries and reinstalling it with another version. Now the code is running.