torch.arange()

TORCH.ARANGE

torch.arange(start=0, end, step=1, *, out=None, dtype=None, layout=torch.strided, device=None, requires_grad=False)→ Tensor

Returns a 1-D tensor of size \left\lceil \frac{\text{end} - \text{start}}{\text{step}} \right\rceil⌈stepend−start​⌉ with values from the interval [start, end) taken with common difference step beginning from start.

Note that non-integer step is subject to floating point rounding errors when comparing against end; to avoid inconsistency, we advise adding a small epsilon to end in such cases.

\text{out}_{{i+1}} = \text{out}_{i} + \text{step}outi+1​=outi​+step

Parameters:

• start (Number) – the starting value for the set of points. Default: 0.

• end (Number) – the ending value for the set of points

• step (Number) – the gap between each pair of adjacent points. Default: 1.

>>> torch.arange(5)
tensor([ 0,  1,  2,  3,  4])
>>> torch.arange(1, 4)
tensor([ 1,  2,  3])
>>> torch.arange(1, 2.5, 0.5)
tensor([ 1.0000,  1.5000,  2.0000])