Torch.cat in 0.4.0

1

In pytorch 0.4.0, how can I concatenate N 0-dimensional tensors to form a single tensor of size N to backprop gradients? In the 0.3.1 version, I can torch.cat() N 1-dimensional tensors to acheieve this, but I wonder how to do this in 0.4.0

2

okay, use torch.stack() instead…

1 Like