search

WongKinYiu / yolov9

Implementation of paper - YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information
GNU General Public License v3.0
9.04k stars 1.44k forks source link

The multi-scale testing issues #351

Open lji72 opened 7 months ago

lji72 commented 7 months ago

I tried to run testing with multi-scale manner and it failed with error. And there is no multi-scale testing guided in readme, so how can I run multi-scale testing with YOLOV9?

JiaminWoo commented 7 months ago

I met the same bug.

And I modified the predict.py according to the _valdual.py as follows:

pred, proto = model(im, augment=augment, visualize=visualize)[:2]
proto = proto[-1]

Now it can test without errors, but I don't know if this use multi-scale for testing.

lji72 commented 7 months ago

@JiaminWoo You can check the accuracy to verify it.