Pytorch Error

[cap-jmk]

When trying to reproduce the examples I recieve the following error when using the mirror flag:

Output

outputFolderolder results/milano/archimboldo/2020-03-14_12-29-31/
Namespace(LS=False, N=80, Ubottleneck=-1, WGAN=True, batchSize=8, beta1=0.5, blendMode=0, cLoss=0, contentPath='samples/archimboldo/', contentScale=1.0, coordCopy=True, dIter=2, fAdvM=0.0, fAlpha=0.1, fContent=1.0, fContentM=1.0, fDiversity=1, fEntropy=0.5, fTV=0.1, firstNoise=False, imageSize=160, lr=0.0002, manualSeed=None, mirror=True, multiScale=False, nBlocks=0, nDep=4, nDepD=4, ndf=80, ngf=80, niter=100, outputFolder='results/milano/archimboldo/2020-03-14_12-29-31/', refine=False, skipConnections=True, testImage='None', texturePath='samples/milano/', textureScale=1.0, trainOverfit=False, workers=0, zGL=20, zLoc=10, zPeriodic=0)
Random Seed:  2593
m1.png img added (1702, 806) total length 1
m2.png img added (1698, 804) total length 2
m4.png img added (1777, 744) total length 3
m3.png img added (1700, 818) total length 4
1200px-Giuseppe_Arcimboldo_-_Self_Portrait_-_Google_Art_Project.jpg img added (1200, 1782) total length 1
image input torch.Size([1, 3, 1760, 1184])
/usr/local/lib/python3.6/site-packages/torch/nn/functional.py:2764: UserWarning: Default grid_sample and affine_grid behavior has changed to align_corners=False since 1.3.0. Please specify align_corners=True if the old behavior is desired. See the documentation of grid_sample for details.
  warnings.warn("Default grid_sample and affine_grid behavior has changed "
tcmalloc: large alloc 2000486400 bytes == 0x555ed222e000 @  0x7fb535602b6b 0x7fb535622379 0x7fb4db742b4a 0x7fb4db7445fa 0x7fb4dd86c78a 0x7fb4ddab530b 0x7fb4ddafcb37 0x7fb526983ad5 0x7fb52698417b 0x7fb526988160 0x7fb52696009b 0x555e58295965 0x555e58205d7b 0x555e582957ce 0x555e582b7cba 0x555e5828ea94 0x555e5828f941 0x555e58295755 0x555e582b8a7a 0x555e58290459 0x555e582911ec 0x555e5830b9a4 0x555e5830bda1 0x555e5830bfa4 0x555e5830fa9e 0x555e581d74be 0x7fb534ffeb97 0x555e582be773
image input torch.Size([1, 3, 806, 1702])
Traceback (most recent call last):
  File "mosaicFAMOS.py", line 69, in <module>
    targetMosaic,templates=getTemplates(opt,N,vis=True,path=opt.outputFolder+desc)
  File "/content/drive/My Drive/Colab Notebooks/SatelliteFAMOS/prepareTemplates.py", line 139, in getTemplates
    out[n:n+1] = randomTile(flow,z)
  File "/content/drive/My Drive/Colab Notebooks/SatelliteFAMOS/prepareTemplates.py", line 96, in randomTile
    f=reflect(f)
  File "/content/drive/My Drive/Colab Notebooks/SatelliteFAMOS/prepareTemplates.py", line 82, in reflect
    return reflect_mirror(x)
  File "/content/drive/My Drive/Colab Notebooks/SatelliteFAMOS/prepareTemplates.py", line 74, in reflect_mirror
    f=torch.where(torch.ByteTensor(f>1), 2-f, f) #so 1:3 maps to 1:-1 - -reflection
TypeError: expected Byte (got Bool)```

[stay-cnsj]

I removed torch.ByteTensor(.) then fixed the error.

[cap-jmk]

Thank you :)

[Daya2821]

When trying to reproduce the examples I receive the following error : I use torch===0.4.1 torchvision===0.4.2.

template size torch.Size([8, 80, 3, 160, 160]) /usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:1698: UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead. warnings.warn("nn.functional.tanh is deprecated. Use torch.tanh instead.") /usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:1709: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead. warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.") Traceback (most recent call last): File "/content/drive/MyDrive/ColabNotebooks/famos-master/mosaicFAMOS.py", line 173, in <module> output = netD(fake.detach())#???why detach File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, **kwargs) File "/content/drive/MyDrive/ColabNotebooks/famos-master/network.py", line 63, in forward output = self.main(input) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, **kwargs) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py", line 119, in forward input = module(input) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, **kwargs) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py", line 399, in forward return self._conv_forward(input, self.weight, self.bias) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py", line 396, in _conv_forward self.padding, self.dilation, self.groups) RuntimeError: set_sizes_and_strides is not allowed on a Tensor created from .data or .detach(). If your intent is to change the metadata of a Tensor (such as sizes / strides / storage / storage_offset) without autograd tracking the change, remove the .data / .detach() call and wrap the change in a with torch.no_grad():` block.
For example, change:
x.data.set_(y)
to:
with torch.no_grad():
x.set_(y)

`