-
Most of IREEs codegeneration relies on using tensor-based approaches. To aid that the `scf.forall` operation allows you to do parallel spatial decomposition. This is some part of what is modeled by `a…
-
This might be a stupid question but I couldn't find a solution anywhere.
When I use gpu to run non-negative decompositions for a random tensor, it is much slower than using a cpu (for various size…
-
Hello,
When I used your tool to analyze my data, I found that the graph of tensor decomposition I ran out had no legend, so I couldn't get which color represented which cell. How did you get the lege…
-
### 📚 The doc issue
Many ops has scalar_tensor version in native_functions.yaml, but are not included in core aten IR. Is it missing or is there some other reason?
### Suggest a potential alternativ…
-
#### Steps or Code to Reproduce
```python
import tensorly as tl
import ivy
tl.set_backend("numpy")
a = tl.ones((2,3,4))
z = tl.decomposition.randomised_parafac(a, 2, 1, n_iter_max=1, init="svd",…
-
Reimplement upsample as trace_only and improve accuracy. In particular, check `aten.upsample_trilinear3d.vec` is accurate.
Related: https://github.com/microsoft/onnxscript/issues/1159
-
Hello,
Thank you for such a wonderful tool! I am trying to run tensor decomposition on my data to compare communication networks in treated vs. untreated samples, but I get the following error I ca…
-
Hi,
I am trying to extend [OverlapPredator.Mink](https://github.com/ShengyuH/OverlapPredator.Mink) to support batch training. Predator is composed of three modules: a Minkowski encoder, a GNN and a…
ghost updated
3 years ago
-
#### Describe the bug
Computation of sparse robust pca crashes as it calls svd with unexpected keyword (`full_matrices`)
#### Steps or Code to Reproduce
Mostly copied from tensorly docs
```pytho…
dawe updated
5 months ago
-
### 🐛 Describe the bug
Repro:
```
import torch
import torch._dynamo
@torch.compile(backend="eager", fullgraph=True)
def fn(x):
return torch.functional.split(x, 0)
fn(torch.empty((0,)…