-
Such linalg.generic can not be vetorized:
```
// mlir-opt test.mlir -transform-interpreter
#map = affine_map (d0, d1, d2, d3)>
#map1 = affine_map (d0, d1, d2, 0)>
module {
func.func @generic_1…
-
## Issue
The current interleaved communication routines in OpenSHMEM (`shmem_iput/iget`) transfer single element chunks that are a fixed stride apart (source and destination can have different stri…
-
Found an repro here:
```
TEST_F(AllocationDomainTest, ReductionOnPermutedReshape) {
auto fusion = std::make_unique();
FusionGuard fg(fusion.get());
TensorView* in = makeContigConcreteTens…
-
This is a boundary case, but important for other modules nonetheless. Define
```
g1 = get_symmetric_group_sgs(0)
g2 = ([], Permutation(1))
g3 = get_symmetric_group_sgs(2)
```
Currently, `g1 = …
-
We probably need to add some more examples and doc for new features in TensorLy.
Would be great for instance to have an example for @earmingol @hmbaghdassarian as a user guide on how to use the Co…
-
Things needed for memory modeling automatic generation of tests
branch name: tiling_loop_generation
file name: scripts/test_generating_code.py
TODO: @Ritvik1sharma needs to split this into test_…
-
### 🚀 The feature, motivation and pitch
Dear PyTorch Devs,
Thank you for your hard work and dedication to creating a great ecosystem of tools and community of users.
This feature request prop…
-
## 🐛 Bug
`x.to(memory_format=torch.contiguous_format)` does not always return a contiguous tensor
## To Reproduce
```
import torch
N, C, H, W = 10, 3, 32, 32
x = torch.empty(N, C, H, W).…
-
Based on discussions in https://github.com/google/heir/issues/906 (e.g., https://github.com/google/heir/issues/906#issuecomment-2292793948), I think it would be prudent to have everyone agree on a not…
j2kun updated
4 weeks ago
-
Thanks for the great work!
I saw the voxelization use reverse_index=True to get coordinate in zyx-format and get dense feature in NCDHW finally. I'm curious about this setting, why use xyz-format?I…