Got stuck at self.G.to(self.device)

[Kirito0816]

Sorry to disturb. I tried to run StarGAN and got this issue. It was stuck here. I search for it and find there was a similar issue on pytorch1.3.0 but not 1.0.0

G The number of parameters: 8430528 Discriminator( (main): Sequential( (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (1): LeakyReLU(negative_slope=0.01) (2): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (3): LeakyReLU(negative_slope=0.01) (4): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (5): LeakyReLU(negative_slope=0.01) (6): Conv2d(256, 512, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (7): LeakyReLU(negative_slope=0.01) (8): Conv2d(512, 1024, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (9): LeakyReLU(negative_slope=0.01) (10): Conv2d(1024, 2048, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1)) (11): LeakyReLU(negative_slope=0.01) ) (conv1): Conv2d(2048, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (conv2): Conv2d(2048, 5, kernel_size=(2, 2), stride=(1, 1), bias=False) ) D The number of parameters: 44762048

Pytorch version is 1.0.0 CUDA version is 8.0 Thank you.

[Kirito0816]

When I try to install pytorch0.4.1 for cuda 8.0, the conda always installs the pytorch for 9.0.176

[Kirito0816]

conda list | grep pytorch cuda80 1.0 h205658b_0 pytorch pytorch 1.0.0 py3.7_cuda8.0.61_cudnn7.1.2_1 [cuda80] pytorch

[Kirito0816]

It turns out that the GPU itself has a problem, not version problem