-
## 🐛 Bug
Error originates from mkl-dnn which uses != operator on thread_id
the != operator on thread_id has been removed in c++20 (was deprecated earlier)
A simpler fix is to do !( thread_id() =…
-
### 🐛 Describe the bug
I am trying to compile PyTorch from source.
```bash
docker run --gpus all -d --name pytorch-build-cuda102 -v /data2:/data2 -w /data2/pytorch-build nvidia/cuda:10.2-cudnn7-d…
-
Intel's DNNL (https://intel.github.io/mkl-dnn/index.html) can boost cpu-only performance up about 35%-40% by simply replacing the "cblas_sgemm" with "dnnl_sgemm".
Intel MKL's "mkl_sgemm" is just a …
-
I am trying to build Tensorflow docker image on an aarch64 device, and it's not able to find the ONEDNN_VERSION (it appears it's blank)
```
Step 36/101 : RUN $PACKAGE_DIR/build-onednn.sh
---> Run…
-
### Prerequisite
- [X] I have searched [Issues](https://github.com/open-mmlab/mmcv/issues) and [Discussions](https://github.com/open-mmlab/mmcv/discussions) but cannot get the expected help.
- [X] Th…
ruckc updated
6 months ago
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. I have read the [FAQ documentation](https://mmdetec…
-
### Checklist
- [x] 1. I have searched related issues but cannot get the expected help.
- [x] 2. The bug has not been fixed in the latest version.
- [x] 3. Please note that if the bug-related iss…
-
please I had following error:
>
Here they are my settings:
System environment:
sys.platform: win32
Python: 3.9.7 (default, Sep 16 2021, 16:59:28) [MSC v.1916 64 bit (AMD64)]
CUDA a…
-
### 🐛 Describe the bug
I'm trying to move clang configs to use clang for all building and linking, but it breaks `MKLDNNs` ability to detect presence of OpenMP on the system, see https://github.com/p…
-
Hi guys,
when i run main.py on my laptop with webcam, it's working but low FPS (1,25 ~ 1,75FPS) and it could not identify the action (just ID and Operate)
And i tried run with video, this issue:
…