Average dectetion rate
You can observe the performance of the model through the average detection rate. The higher the average detection rate, the better the performance of the model.
After training the annotated images, the model is used to detect these images again, and the average detection rate is obtained by comparing the annotation results and the prediction results.
The average detection rate only represents the performance of the model on annotated images. If the test images differ significantly from the annotated images, the detection performance may not be good.
If the average detection rate is low, the offline test result will be poor. Please increase the number of annotated images to improve the average detection rate.
Still have a question?
Each section should be concise, user-friendly, and direct users to additional resources or documentation when necessary.