-
Hello,
I'm getting the following problem, when I run the command ```python -m nuitka --follow-imports --standalone .\app-demo.py``` on my project.
```
Nuitka-Options:INFO: Used command line optio…
-
Hello,
I'm getting the following problem, when I run the command ```python -m nuitka --follow-imports --standalone .\app-demo.py``` on my project.
```
Loading model
INFO:root:Loaded coca_ViT-L-1…
-
我在测试通过 fleet 在 "intel_gpu" 单机多卡的机器上训练 DDP RN50。forward 正常，backward fail 在一个错误的 memcpy 上：
```
Copy 2048 Bytes from 0xffff8181ff0cc800(Place(cpu)) to 0x1ab0f000(Place(cpu))
```
这里的 src 显然是一个 device …
-
### 🐛 Describe the bug
hi，when I compile thirty party collective communication library for pytorch 1.13 supported by NGC images "nvcr.io/nvidia/pytorch:22.11-py3", I find that some head files don't…
-
Hello,
I noticed that there is correctness issue on CCL test with shm provider. This test starts bunch of allreduce operations in parallel, then complete them all and then do correctness check.
Wi…
-
I am running a few CNN/DNN examples within oneDNN, but it always run on a single Nvidia GPU. Is there a way to run on multiple Nivida GPUs? Is it currently supported or do I miss something? Thanks!
-
## 🚀 Feature
This issue is for discussing the integration of the Sequential Aggregation and Rematerialization technique for distributed full-graph training and inference from [here](https://github.co…
-
### Description of the problem
I try to move a machine translation inference application which base on pytorch to gramine-sgx, for protect the translation process.
After run in gramine-sgx succe…
-
Hello,
I'm getting the following problem, when I run the compiled output of my project, which includes the transformers library.
```
None of PyTorch, TensorFlow >= 2.0, or Flax have been found. M…
-
I built oneccl_bind_pt from Source and set 'COMPUTE_BACKEND' to 'dpcpp' when built it. I found intel_extension_for_pytorch only v1.10.200 was released with XPU support, so I installed it with torch 1.…