-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
I'm trying to implement a single kernel which performs matrix multiplication followed by a non-unary activation function f(x, y) = sigmoid(x)*tanh(y). In a neural network library this would implement…
-
## 🐛 Bug
Matrix multiplication does not work properly on Torch 1.8.1 with CUDA 11.1 when running on a 1080Ti with 460 or 465 Nvidia drivers.
## To Reproduce
1. Save this test script as test.p…
-
**Describe the bug**
The Raspberry Pi 5 4GB performs slightly better (0-10%) than the 8GB version at default 2.4 GHz and the gap widens to >100% for certain workloads when overclocked. These workload…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no si…
-
由于python版本要求使用3.8版本,不能直接使用安装python3.10版本的wheel包
paddle_custom_mlu.whl
可以给出paddlecustomdevice源码编译的步骤和命令么?谢谢!
@YanhuiDua
-
Just tested it in IPython
```
import torch as t
conv2d = t.nn.Conv2d(32,32,3,1,1).cuda()
conv2d_depthwise = t.nn.Conv2d(32,32,3,1,1,groups=32).cuda()
inp = t.randn(2,32,512,512).cuda()
# w…
-
While executing the training script. I encountered the following error.
```bash
Traceback (most recent call last):
File "train.py", line 72, in
model.optimize_parameters()
File "/home/D…
-
I have built openmpi 4.0.1 against UCX 1.5.2 (and also 1.6) and get segmentation faults in libucs in mpi4py when it is compiled against this MPI. Here are my configure flags for UCX and OpenMPI:
``…
-
Not sure if I'm doing something illegal here, but I get an assertion error from the following set of llops
```julia
using LoopVectorization
function avxbug(U = randn(2,2), E1 = randn(2))
t =…