Slow Training on Jetson Orin

[LinirZamir]

Search before asking

• [X] I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

I am currently trying to train a very simple model on my Jetson Orin.

The purpose is to train it using custom dataset of only 10 images, for 100 epochs. I am using the yolov5/train.py with a custom dataset and yaml file.

When I train it on my Lenovo Titan Nvidia, it runs fast and the results are good. However, when I test it on a newly flashed Jetson Orin with Pytorch version 1.13 and Cuda 11.4 it runs extremely slow..

It must be something to do with the configuration, but I couldn't find any possible solutions to this issue. Training command:

python3 yolov5/train.py --data custom.yaml

Also I am getting a WARNING NMS time limit 0.850s exceeded every epoch.

I would appreciate any help on the matter. Thank you.

Additional

No response

[github-actions[bot]]

👋 Hello @LinirZamir, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email support@ultralytics.com.

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training, validation, inference, export and benchmarks on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[github-actions[bot]]

👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs.

Access additional YOLOv5 🚀 resources:

• Wiki – https://github.com/ultralytics/yolov5/wiki
• Tutorials – https://docs.ultralytics.com/yolov5
• Docs – https://docs.ultralytics.com

Access additional Ultralytics ⚡ resources:

• Ultralytics HUB – https://ultralytics.com/hub
• Vision API – https://ultralytics.com/yolov5
• About Us – https://ultralytics.com/about
• Join Our Team – https://ultralytics.com/work
• Contact Us – https://ultralytics.com/contact

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐!

[glenn-jocher]

@LinirZamir hi there,

Training YOLOv5 on Jetson Orin can indeed be slower compared to high-end desktop GPUs due to hardware limitations. You might want to consider reducing the model size or batch size to speed up training. Additionally, the NMS time limit warning can indicate that the NMS (Non-Maximum Suppression) is taking too long, which could also affect training speed.

Make sure you have installed the necessary dependencies for GPU support on Jetson Orin and consider performance optimization strategies such as adjusting batch size and model architecture. For more detailed guidance, check out the YOLOv5 documentation at https://docs.ultralytics.com/yolov5/.

Hope this helps!