search

ultralytics / yolov5

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
47.78k stars 15.73k forks source link

How to use tensor rt in yolov5 detection #12973

Open tasyoooo opened 1 month ago

tasyoooo commented 1 month ago

Search before asking

Question

I'm using local machine with gpu rtx3050. I would like to utilize my gpu during the detection process. I am using webcam as a soure and framework tensor rt

Additional

No response

glenn-jocher commented 1 month ago

Hello there! 👋

Great to hear you're leveraging YOLOv5 with TensorRT for improved performance on your RTX3050 GPU! Using TensorRT, you can significantly speed up inference time by optimizing neural network models.

Here's a general overview of the steps involved:

  1. Export YOLOv5 Model to ONNX: Convert your trained YOLOv5 model to ONNX format. You can do this with the export.py script in the YOLOv5 repository.
python export.py --weights yolov5s.pt --img 640 --batch 1 --device 0 --opset 12 --include onnx
  1. Convert ONNX Model to TensorRT Engine: Use the trtexec command or TensorRT Python API to convert the ONNX model to a TensorRT engine optimized for your GPU.
trtexec --onnx=yolov5s.onnx --saveEngine=yolov5s.engine
  1. Perform Inference with TensorRT: Finally, you can load the TensorRT engine and perform inference. You'll need to handle pre-processing of your webcam feed and post-processing of the detection outputs according to YOLOv5's requirements.

While the above steps provide a high-level overview, specific implementation details can vary. For further guidance, checking documentation and examples specific to TensorRT and YOLOv5 is recommended. Feel free to explore our official documentation for more insights: https://docs.ultralytics.com/yolov5/

Wishing you success in your project! If you have any more questions, feel free to ask. 🚀

github-actions[bot] commented 3 days ago

👋 Hello there! We wanted to give you a friendly reminder that this issue has not had any recent activity and may be closed soon, but don't worry - you can always reopen it if needed. If you still have any questions or concerns, please feel free to let us know how we can help.

For additional resources and information, please see the links below:

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 YOLO 🚀 and Vision AI ⭐