Replies: 1 comment 1 reply
-
That is the regular dehavior. Some layers are popped from the classifiers in order to use them as feature extractors. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Although this may make me appear ignorant, I truly hope that it will be possible to execute the yolov8_tracking project in Windows in the future. When I attempted to run the project in Windows, the result that popped up was:
The following layers are discarded due to unmatched keys or layer size: ['classifier.weight', 'classifier.bias']
Beta Was this translation helpful? Give feedback.
All reactions