-
### Your current environment
```text
The output of `python collect_env.py`
```
PyTorch version: 2.3.0
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
O…
-
### Your current environment
```text
PyTorch version: 2.3.1+cpu
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A
OS: Ubuntu 24.04 LTS (x86_64)
GCC vers…
-
### 🐛 Describe the bug
I exported the same model twice but the hash results are not same, for `torch-2.6.0.dev20241029+cu121`. I expect `hash1` equals `hash2`. The following function failed for `to…
-
A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.0.1 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need…
-
### 🐛 Describe the bug
Same data all reduce on H20 and tp8, but the results are different.
The problem can be reproduced on the image：nvcr.io/nvidia/pytorch:24.09-py3
demo code：
``` python
…
-
### 🐛 Describe the bug
Using nested tensors generated with `torch.narrow` as inputs to `torch.nn.functional.scaled_dot_product_attention` works fine in the forward pass of the model. However, both …
-
Possible duplicate of #5, with [solution](https://github.com/user-attachments/files/16560669/install_dependencies.zip) provided there already, but I'm going to post an issue ticket here as a warning. …
-
Hi, I lack experience with all of this tooling. After following the README I've run the FluxPipeline example, and it appears to hang. I was anticipating that it would complete and write the file to …
-
### Your current environment
Collecting environment information.
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
0S: Cent0S Li…
-
### 🐛 Describe the bug
Just installed pytorch nightly on my Arc 770 using
```
pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu
```
The…