-
### 🐛 Describe the bug
The goal is to get all of these working:
```
xfail('fft.fft2', ''), # aten.size.default - couldn't find symbolic meta function/decomposition
xfail('fft.fft', ''…
-
While I am aware of the libcudnn problem, the training continued but at the end I got the error you see below. All the steps before this step went through with no problem. Please let me know what your…
-
## Page 14
- [x] The progress of computer simulation gradually ~allows the virtual resolution~ _solves_ **Virtually veut dire presque en anglais**
- [x] to pursue these studies ~for~ _at_ a reason…
-
**Submitting author:** @kokitsuyuzaki (Koki Tsuyuzaki)
**Repository:** https://github.com/rikenbit/nnTensor
**Branch with paper.md** (empty if default branch):
**Version:** v1.1.14
**Editor:** @prash…
-
May I ask what should be the correct dimension of grads?
Suppose the output feature map size is 100, and the batch_size is 512, should the rep and grads be of size [512, 100]? And how do you deal wi…
-
Click to expand!
### Issue Type
Bug
### Have you reproduced the bug with TF nightly?
Yes
### Source
source
### Tensorflow Version
2.8.0
### Custom Code
No
### OS Platform and Distributio…
-
In doing annotation over a large library I found myself unable to annotate some matrix operations fully. For example, the library contains an SVD function and the output size depends on the MIN of t…
-
```haskell
alignWithList :: Traversable t => (Maybe a -> b -> c) -> [a] -> t b -> t c
alignWithList f xs = snd . mapAccumL go xs
where go [] y = ([], f Nothing y)
go (x:xs) y = (xs, …
-
## Issue description
The following code produces a dispatch error in functionalization:
```python
import torch
def fn(x):
p = torch.nn.Parameter(x)
return p[0]
if __name__ == '_…
-
**Is your feature request related to a use case or problem? Please describe.**
As part of providing tools for allocating / deallocating qubits in https://github.com/quantumlib/Cirq/issues/6040, we wa…