I’ve created an object detection model for YOLOv8, and I’m curious about what the values in the images represent. Also, I’d like to know whether the values are taken from the training set or the test set when generating the validation.
Hi!
(Keep in mind that I’m not a Roboflow Employee but this is my knowledge on the subject 
)
Your model tracks the Precision (how often the model is correct in its prediction) of a crack/pothole/all (both combined). Therefore, the higher the number, the better as the model is more often correct. It tracks cracks better than potholes in terms of precision, and it tracks potholes and both combined equally (or around equally).
The validation Set’s values are gotten specifically from the training set as it is used to fine-tune the model to make it more accurate. The testing set tests the model, the Training set trains the model, Validation set makes the model better through fine-tuning (during the training phase).
Video 1 could be helpful for more information on Mean Average Precision
Video 2 could be helpful for more information on Validation
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.