-
The hashing currently uses webui's old `sd_models.model_hash()`, which is prone to collisions
Just the tensor part of the file should be hashed instead, and the length of the hash increased somewha…
-
I have modified alexnet example to use my own dataset. So training works fine. However, I want to test the accuracy so I am trying to use the model.predict() and model.evaluate() methods.
I have crea…
-
Repro:
```py
import torch
class DiffTwoTensor(torch.Tensor):
def __new__(
cls,
data: torch.Tensor,
):
self = torch.Tensor._make_wrapper_subclass(
…
-
### 🐛 Describe the bug
Compare this meta implementation to the one registered in native_functions.yaml. Meta implementation is missing the `non_blocking` argument.
https://github.com/pytorch/pytorch…
-
`compose(ATensorbacked.class).from(aTensor);`
vs.:
`aTensor.as(ATensoribacked.class);`
(usually, the latter one should be possible ... as long it does not require anything additional (e.g. a field)…
-
TensorFlow may use ByteBuffers directly for creating the tensors while other buffers are copied.
From the [Documentation of the Tensor class](https://www.tensorflow.org/api_docs/java/reference/org/…
-
Hello! I'm a beginner to learning this amazing work.
I want to ask 2 questions:
1.How to recover a tensor from bytes? I saw this example:
![image](https://user-images.githubusercontent.com/48152144…
-
## 🚀 Feature
Currently `ReflectionPadNd` does *not* repeat the boundary pixels, as demonstrated by the docs:
```
>>> m = nn.ReflectionPad2d(2)
>>> input = torch.arange(9, dtype=torch.float).re…
-
### Describe the issue:
These are as expected:
```
from collections,abc import Iterable
>>> isinstance(pt.constant([1,2,3]), Iterable)
True
>>> isinstance(pytensor.shared(np.array([1,2,3])), I…
-
Hi, thanks for sharing your research.
I need to stack vertices and faces in a batch, but I note that all your experiments in the repository have batch size=None. Have you tried stacking vertices an…