Alan Zhai

cant download model and json

10

for branch release/2.1 https://github.com/PaddlePaddle/PaddleNLP/blob/0e1f3682f8038b0f93b15565acb4b2edbb5d9a1e/paddlenlp/transformers/ernie_gen/modeling.py#L268 some link has gone. could anyone fix it? Thanks.

much thanks to your helpful tutorial. Could you help provide a lecture about how to improve kernel perf using cuda perf tools? And Any other link or info of ECE-508...

[ONNX] support export onnx format

3

### Is your feature request related to a problem? Please describe. There is no any onnx export solution for deployment. ### Solutions Such repo is under pytorch development while some...

How to switch different doc version?

6

https://oneapi-src.github.io/oneDNN/ is always the latest docs. If I want to read the old one, how to switch it?

sysmon failed on B580

17

## Reproduce - OS: ubuntu 24.04 - kernel: 6.13.4 - compute runtime: https://github.com/intel/compute-runtime/releases/tag/25.05.32567.17 - dGPU: B580 ```bash dpkg -l | grep intel ii intel-gsc 0.9.5-112~u24.04 amd64 Intel(R) Graphics System Controller...

1. two Arc GPUs on ubuntu OS 2. write a sycl vector-add code and bind queue to the device GPU:0 3. run the program and monitor by `zesDeviceProcessesGetState` 4. `zesDeviceProcessesGetState`...

Please privide more 4bit matmul docs and example?

10

oneDNN enables 4bits for matmul, like u4/s4 and float-4bit. However, there is no document or example about 4-bit data type. Especially, 4bit storage structure and use cases in the real...