RuntimeError: std::exception

vision
1

Hi everyone, I am running the Discriminator from here with ndf = 32 and I am getting the following error:

Traceback (most recent call last):
  File "main.py", line 152, in <module>
    train(epoch, model, opt, dataset_train, train_logger)
  File "main.py", line 88, in train
    model.train_step(imgs_, labels)
  File "/home/darley/Documents/Repositories/attention-based-classification/models/ad_attn_vae.py", line 119, in train_step
    output = self.disc(same_class).view(-1)
  File "/home/darley/miniconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/darley/Documents/Repositories/attention-based-classification/models/disc.py", line 51, in forward
    output = self.conv5(output)
  File "/home/darley/miniconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/darley/miniconda3/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 320, in forward
    self.padding, self.dilation, self.groups)
RuntimeError: std::exception

My Pytorch cpu version is 1.0.0 from conda on python 3.7.1. This error is happening at the last conv layer nn.Conv2d(ndf * 8, 1, 4, 1, 0, bias=False).
What does this error mean?

2

I found the error. Before the last Conv2d, my tensor had shape torch.Size([8, 128, 2, 2]), then when it goes to the last Conv2d it crashes. So I am not using the two last convs.