pytorch view 1 meaning

1import torch
2
3x = torch.arange(6)
4
5print(x.view(3, -1))      # inferred size will be 2 as 6 / 3 = 2
6# tensor([[ 0.,  1.],
7#         [ 2.,  3.],
8#         [ 4.,  5.]])
9
10print(x.view(-1, 6))      # inferred size will be 1 as 6 / 6 = 1
11# tensor([[ 0.,  1.,  2.,  3.,  4.,  5.]])
12
13print(x.view(1, -1, 2))   # inferred size will be 3 as 6 / (1 * 2) = 3
14# tensor([[[ 0.,  1.],
15#          [ 2.,  3.],
16#          [ 4.,  5.]]])
17
18# print(x.view(-1, 5))    # throw error as there's no int N so that 5 * N = 6
19# RuntimeError: invalid argument 2: size '[-1 x 5]' is invalid for input with 6 elements
20
21print(x.view(-1, -1, 3))  # throw error as only one dimension can be inferred
22# RuntimeError: invalid argument 1: only one dimension can be inferred
23

1If there is any situation that you don't know how many rows you want but are sure of the number of columns, then you can specify this with a -1. (Note that you can extend this to tensors with more dimensions. Only one of the axis value can be -1). This is a way of telling the library: "give me a tensor that has these many columns and you compute the appropriate number of rows that is necessary to make this happen".