-
### Your current environment
The output of `python collect_env.py`
```text
python collect_env.py
Collecting environment information...
2024-09-23 17:57:46.577274: I tensorflow/core/util/po…
-
### 🐛 Describe the bug
import torch
x = torch.tensor([[757.5]])
y = torch.quantize_per_channel(x, torch.tensor([15.0]), torch.tensor([0]), 0, torch.qint8).int_repr()
print(y)#51
real: torch.r…
-
### 🐛 Describe the bug
I ran into a counterintuitive behavior that might be a result of the interplay between strides, Conv2d, and tf32. The code to reproduce the behavior is given below:
```pyt…
-
### Your current environment
```
PyTorch version: 2.1.2+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 vers…
-
### 🐛 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)
``…
-
### This issue respects the following points:
- [X] This is a **bug**, not a question or a configuration issue; Please visit our forum or chat rooms first to troubleshoot with volunteers, before crea…
-
### 🐛 Describe the bug
We are facing issues with loss curves and reproducibility when using `torch.compile()` with our models. Attached below is a graph of train loss with runs with `torch.compile(…
-
### What is the issue?
OS: Ubuntu 24.04 LTS
GPU: Nvidia Tesla P40 (24G)
I installed ollama without docker and it was able to utilise my gpu without any issues.
I then deployed ollama using the f…
-
# 🐛 Bug
Reading Cuda tensors from multiprocessing queue causes child (reader) process to hang.
I discovered that process hangs on any operation (sum, min, max, mean, etc.), in my example code on "…
-
### Your current environment
The output of `python collect_env.py`
```
Collecting environment information...
PyTorch version: N/A
Is debug build: N/A
CUDA used to build PyTorch: N/A
ROCM us…