-
#### Is your feature request related to a problem? Please describe.
I want to correct for missing values using Tucker or CP decomposition in data containing categorical data.
#### Describe the s…
-
### 🐛 Describe the bug
`torch.compile(backend=toy_backend, dynamic=True)(model)` can't trace the shape on 3DRESET correctly. Some tensor have empty shape `[]`
colab reproduce: https://colab.resear…
-
Hi,
Thank you for this library!
I am wondering if there is an easy way to extract the Fourier modes being learnt. I am hoping to extract the state of the tensor, x, immediately after the inverse…
-
Kinda weird.
cc @ezyang
-
PyTorch Model:
```python
class NeuralNet(torch.nn.Module):
def __init__(self, input_size, hidden_size, num_classes):
super().__init__()
self.fc1 = torch.nn.Linear(input_si…
-
When running onnx dynamo export
```python
from torch import nn
import torch
class Model(nn.Module):
def __init__(self):
super().__init__()
self.emb = nn.Embedding(15, 96…
-
### 🐛 Describe the bug
Greetings,
I have been trying to get `torch.compile` to work with a Mask R-CNN model, but have not been able to do so in combination with gradient disabling. An error also o…
-
### 🐛 Describe the bug
This is similar to #92324 .
Program to reproduce:
```python
import torch
def fn(x):
return torch.bitwise_xor(x, 18)
# return torch.bitwise_xor(x, torch.tens…
-
### 🐛 Describe the bug
_No response_
### Error logs
~~~
Traceback (most recent call last):
File "/scratch/anijain/work/pytorch/repro.py", line 52, in
res = run_fwd_maybe_bwd(opt_mod, args…
-
Hi 1dnn team,
I was hoping to get some confirmation or guidance about some behavior I'm seeing when running a separate network per thread. Specifically, for an 4 physical core machine with hyperthr…