-
### 🐛 Describe the bug
# install CPU version
pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cpu
# verify torch
python -c 'import tor…
nctu6 updated
2 months ago
-
### Your current environment
PyTorch version: 2.4.1+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (U…
-
### Your current environment
Collecting environment information...
PyTorch version: 2.3.0
Is debug build: False
CUDA used to build PyTorch: 12.2
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.0…
-
### 🐛 Describe the bug
For `TransformerEncoder` network, when I created a forward hook, it does not get called if there are `model.eval()` and with `torch.no_grad():` at the same time. Here is a code…
-
### 🐛 Describe the bug
I’ve been working on integrating libtorch C++ into my project for a few weeks and while it works fine when I include the
```
#include
#include
```
headers in a cpp sou…
-
### 🐛 Describe the bug
```python
import torch
x = torch.tensor([2**-10], dtype=torch.float16, device='cuda')
exp = torch.tensor([20], dtype=torch.float16, device='cuda')
torch.ldexp(x, exp)
``…
-
### 🐛 Describe the bug
Since I upgraded torch from `1.13.0+cu117` to `2.0.0+cu117`, the following code isn't logging nor printing the stack trace.
```python
import torch
from torch.profiler im…
-
### 🐛 Describe the bug
ONNX-exporting a PyTorch model which includes the Torch functional `interpolate` in mode `'bilinear'` and `antialias=True` fails, both when using the old ONNX export functional…
-
### 🐛 Describe the bug
DTensor random numbers provide an offset based RNG state tracker OffsetBasedRNGTracker for CUDA. However, for CPU, this offset based RNG state tracker is not available, and it …
-
### 🐛 Describe the bug
I run the same code in CUDA 11.8 and 12.1 and got different results:
```python
import torch
import numpy
from torch import nn
torch.manual_seed(0)
conv = nn.Conv2d(…