I have a problem with the YOLOv5 since torch not detect GPU, the application run on CPU and have terrible performance. I have researched other topics in this forum but all related to CUDA not detect and not help me.
Thank you in advance.
Now YOLOv5 can detect and use the GPU by install the PyTorch follow the instruction ( Won’t detect GPU,CUDA if install pytorch from pip )
PyTorch for Jetson
with the newest Pytorch 1.13
mismatch the torch and torchvision, I also follow
GitHub - pytorch/vision: Datasets, Transforms and Models specific to Computer Vision
get torch 1.12.0 and torchvision 1.12.0 but not help me. So I update both of them to lasted version 1.13 but not help too
Nvidia member can help me to solve this problem to make the comprehensive setup guideline for new user ^^.
Our user reports that the combination of PyTorch nv22.09 + TorchVision 0.15.0 is working.
Could you please give it a try?
I think I have resolved the problem.
Today 05/10/2022 Nvidia has uploaded a new version of Torch+CUDA support compatible with Jetpack 5.0.2.
So I have installed the last one and I have build Torchvision from source here.
After doing that, I have Torch and TorchVision both with CUDA support I think.
I tried and ran the val.py scipt from yolov5 and it worked. Now I wil try to run it using a TensorRT engine, I hope there will be no issues. I will keep you updated.
More, we also have a container that includes PyTorch and TorchVision together.
Get started on your AI journey quickly on Jetson. The Machine learning container contains TensorFlow, PyTorch, JupyterLab, and other popular ML and data science frameworks such as scikit-learn, scipy, and Pandas pre-installed in a Python 3.6...
The first solution work for me. Thank you for point out solution ^^
Good to know it works.
Thanks for the update.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.