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

yolov8 的结果和预期不一致,只有macOS(M1)受影响 #5340

Closed
kekxv opened this issue Feb 20, 2024 · 3 comments
Closed

yolov8 的结果和预期不一致,只有macOS(M1)受影响 #5340

kekxv opened this issue Feb 20, 2024 · 3 comments

Comments

@kekxv
Copy link

kekxv commented Feb 20, 2024

正确输出了结果,但是有偏差;关闭 fp16 之后结果正常

https://github.com/Tencent/ncnn/releases/tag/20230517 以及后续出现错误结果
https://github.com/Tencent/ncnn/releases/tag/20230223 结果正确

how to reproduce | 复现步骤 | 再現方法

  1. 使用 yolov8 导出的模型,或者用 https://github.com/kekxv/YOLOv8Tutorial/releases/tag/20240219 里面的测试

错误的结果:
截屏2024-02-20 22 45 11
正确的结果:
截屏2024-02-20 22 44 52

@kekxv
Copy link
Author

kekxv commented Feb 21, 2024

受影响开始的提交 :
010d677

像是 softmax_arm.cpp 部份

nihui added a commit to nihui/ncnn that referenced this issue Mar 28, 2024
@nihui
Copy link
Member

nihui commented Mar 28, 2024

#5393

@nihui nihui closed this as completed in 167501f Mar 30, 2024
senli123 pushed a commit to senli123/ncnn that referenced this issue May 7, 2024
@nihui
Copy link
Member

nihui commented Jan 8, 2025

hi, yolov8 examples are updated with full support for detection, segmentation, classification, pose estimation and obb
https://github.com/Tencent/ncnn/tree/master/examples

android demo
https://github.com/nihui/ncnn-android-yolov8

detailed instruction (zh)
https://zhuanlan.zhihu.com/p/16030630352

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