Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Can Darknet run YOLOv7 and YOLOv7-tiny? #1877

Closed
NickDatLe opened this issue Oct 9, 2023 · 2 comments
Closed

Can Darknet run YOLOv7 and YOLOv7-tiny? #1877

NickDatLe opened this issue Oct 9, 2023 · 2 comments

Comments

@NickDatLe
Copy link

I would like to run YOLOv7-tiny on Darknet if it is faster than the pytorch implementation. So my questions are:

  1. Can yolov7 run on darknet?
  2. Can yolov7-tiny run on darknet?
  3. Is Darknet faster than pytorch? (Would be the only practical reason I would use Darknet over pytorch)
@SheldonWBM
Copy link

SheldonWBM commented Oct 10, 2023

  1. Yes, read the documentation:
    https://github.com/pjreddie/darknet
    Official YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors pjreddie/darknet#2557
  2. Yes, you can use yolov7, yolov7x and yolovy7-tiny
  3. For training, probably not any faster. At runtime (i.e. actively detecting objects in a live stream) C/C++ is faster (higher FPS) than Python (from experience). Darknet can run in C/C++ or Python mode (read documentation). If you are changing OpenCV, please make sure you modify the appropriate OpenCV for your runtime environment. Note, if you modify C/C++ code, you must compile it to see changes. Python does not require compiling before running which, might be one reason it is slower.

@NickDatLe
Copy link
Author

  1. detectors Official YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors pjreddie/darknet#2557

Thank you so much @SheldonWBM for your answers. I will test darknet to run inference.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants