frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Pointer to the nonzero limit ticket (#124244)

Browse Source 
For the nonzero impl limits we are still asking at runtime to fill a new ticket  but we had already more then one.
So I am pointing to the current open ticket.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/124244
Approved by: https://github.com/ezyang
This commit is contained in:
bhack
2024-04-17 06:15:32 +00:00
committed by PyTorch MergeBot
parent dd3cea3291
commit ed22dde877
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
aten/src/ATen/native/cuda/Nonzero.cu
View File

@ -112,7 +112,7 @@ void nonzero_cuda_out_impl(const Tensor& self, Tensor& out){
Tensor& nonzero_out_cuda(const Tensor& self, Tensor& out){
TORCH_CHECK(self.numel() < std::numeric_limits<int>::max(), "nonzero is not supported for tensors with more than INT_MAX elements, \
file a support request");
See https://github.com/pytorch/pytorch/issues/51871");
TORCH_CHECK(out.dtype() == at::kLong, "Expected object of scalar type ", at::kLong, " as out, but got ", out.dtype());
TORCH_CHECK(self.device() == out.device(), "expected self and out to be on the same device, but got out on ",
out.device(), " and self on ", self.device());

2
aten/src/ATen/native/mps/operations/Indexing.mm
View File

@ -266,7 +266,7 @@ Tensor& nonzero_out_mps(const Tensor& self, Tensor& out_) {
TORCH_CHECK(self.numel() < std::numeric_limits<int>::max(),
"nonzero is not supported for tensors with more than INT_MAX elements, \
file a support request");
See https://github.com/pytorch/pytorch/issues/51871");
TORCH_CHECK(
out_.dtype() == at::kLong, "Expected object of scalar type ", at::kLong, " as out, but got ", out_.dtype());
TORCH_CHECK(self.device() == out_.device(),