torch.nonzero¶
- torch.nonzero(input, *, out=None, as_tuple=False) LongTensor or tuple of LongTensors ¶
Note
torch.nonzero(..., as_tuple=False)
(default) returns a 2-D tensor where each row is the index for a nonzero value.torch.nonzero(..., as_tuple=True)
returns a tuple of 1-D index tensors, allowing for advanced indexing, sox[x.nonzero(as_tuple=True)]
gives all nonzero values of tensorx
. Of the returned tuple, each index tensor contains nonzero indices for a certain dimension.See below for more details on the two behaviors.
When
input
is on CUDA,torch.nonzero()
causes host-device synchronization.When
as_tuple
isFalse
(default):Returns a tensor containing the indices of all non-zero elements of
input
. Each row in the result contains the indices of a non-zero element ininput
. The result is sorted lexicographically, with the last index changing the fastest (C-style).If
input
has dimensions, then the resulting indices tensorout
is of size , where is the total number of non-zero elements in theinput
tensor.When
as_tuple
isTrue
:Returns a tuple of 1-D tensors, one for each dimension in
input
, each containing the indices (in that dimension) of all non-zero elements ofinput
.If
input
has dimensions, then the resulting tuple contains tensors of size , where is the total number of non-zero elements in theinput
tensor.As a special case, when
input
has zero dimensions and a nonzero scalar value, it is treated as a one-dimensional tensor with one element.- Parameters:
input (Tensor) – the input tensor.
- Keyword Arguments:
out (LongTensor, optional) – the output tensor containing indices
- Returns:
If
as_tuple
isFalse
, the output tensor containing indices. Ifas_tuple
isTrue
, one 1-D tensor for each dimension, containing the indices of each nonzero element along that dimension.- Return type:
LongTensor or tuple of LongTensor
Example:
>>> torch.nonzero(torch.tensor([1, 1, 1, 0, 1])) tensor([[ 0], [ 1], [ 2], [ 4]]) >>> torch.nonzero(torch.tensor([[0.6, 0.0, 0.0, 0.0], ... [0.0, 0.4, 0.0, 0.0], ... [0.0, 0.0, 1.2, 0.0], ... [0.0, 0.0, 0.0,-0.4]])) tensor([[ 0, 0], [ 1, 1], [ 2, 2], [ 3, 3]]) >>> torch.nonzero(torch.tensor([1, 1, 1, 0, 1]), as_tuple=True) (tensor([0, 1, 2, 4]),) >>> torch.nonzero(torch.tensor([[0.6, 0.0, 0.0, 0.0], ... [0.0, 0.4, 0.0, 0.0], ... [0.0, 0.0, 1.2, 0.0], ... [0.0, 0.0, 0.0,-0.4]]), as_tuple=True) (tensor([0, 1, 2, 3]), tensor([0, 1, 2, 3])) >>> torch.nonzero(torch.tensor(5), as_tuple=True) (tensor([0]),)