-
### 1. Issue or feature description
Failing to start HW Accelerated containers.
```
kyle@bently 03:50:18 /var/log $ sudo docker run --gpus all nvidia/cuda:11.3.0-runtime-ubuntu20.04 nvidia-smi
d…
-
### NVIDIA Open GPU Kernel Modules Version
nvidia-open-dkms 530.41.03-3
### Does this happen with the proprietary driver (of the same version) as well?
I cannot test this
### Operating Sys…
-
### 🐛 Describe the bug
When `torch.einsum` evaluates a given expression, the strategy it chooses depends on the hardware it's running on but this strategy selection doesn't (correctly) consider memor…
-
### 🐛 Describe the bug
When enabling amp training with PyTorch native autocast, I noticed there seems to be obvious difference for DDP based model and FSDP based model.
Here is a minimum example …
-
### 🐛 Describe the bug
The problem only occurs under some specific conditions:
1. The video backend is 'video_reader'
2. The VideoReader object is initialized with raw bytes rather than a path stri…
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
PyTorch version: 2.5.1+cpu
Is debug build: False
CUDA used to build PyTorch: …
-
If I apply your repo, can I divide the GPU like this and check it on nvidia-smi?
like this: https://docs.run.ai/Researcher/Walkthroughs/walkthrough-fractions/
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
PyTorch version: 2.4.0+cu121
Is debug build: False
CUDA used to build PyTorch…
-
Hello,
My company is currently studying the possibility to use vJoule to measure the energy consumption of some of our applications, but unfortunately we are not able to use it because all our server…
-
## Description
When adding NDArray on different contexts, I get either:
- warning of different context: GPU 0 -> CPU
- error + crash: GPU 0 -> GPU N with N != 0
## Environment info (Required)
…