I'm struggling with understanding the way torch.permute() works. In general, how exactly is an n-D tensor permuted? An example with explaination for a 4-D or higher dimension tensor is highly appreciated. I've search across the web but did not find any clearly explaination. Thank you! Continue reading...
